2024
Cayetano, Arjay; Stransky, Christoph; Birk, Andreas; Brey, Thomas
Fish Age Reading using Deep Learning Methods for Object-Detection and Segmentation Journal Article
In: ICES Journal of Marine Science, vol. 81, no. 4, pp. 687–700, 2024.
@article{DL-fish-age-reading-otolithes-JMS24,
title = {Fish Age Reading using Deep Learning Methods for Object-Detection and Segmentation},
author = {Arjay Cayetano and Christoph Stransky and Andreas Birk and Thomas Brey},
url = {https://academic.oup.com/icesjms/article/81/4/687/7614790},
doi = {10.1093/icesjms/fsae020},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {ICES Journal of Marine Science},
volume = {81},
number = {4},
pages = {687–700},
abstract = {Determination of individual age is one essential step in the accurate assessment of fish stocks. In non-tropical environments, the manual count of ring-like growth patterns in fish otoliths (ear stones) is the standard method. It relies on visual means and individual judgment and thus is subject to bias and interpretation errors. The use of automated pattern recognition based on machine learning may help to overcome this problem. Here, we employ two deep learning methods based on Convolutional Neural Networks (CNNs). The first approach utilizes the Mask R-CNN algorithm to perform object detection on the major otolith reading axes. The second approach employs the U-Net architecture to perform semantic segmentation on the otolith image in order to segregate the regions of interest. For both methods, we applied a simple postprocessing to count the rings on the output masks returned, which corresponds to the age prediction. Multiple benchmark tests indicate the promising performance of our implemented approaches, comparable to recently published methods based on classical image processing and traditional CNN implementation. Furthermore, our algorithms showed higher robustness compared to the existing methods, while also having the capacity to extrapolate missing age groups and to adapt to a new domain or data source.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Determination of individual age is one essential step in the accurate assessment of fish stocks. In non-tropical environments, the manual count of ring-like growth patterns in fish otoliths (ear stones) is the standard method. It relies on visual means and individual judgment and thus is subject to bias and interpretation errors. The use of automated pattern recognition based on machine learning may help to overcome this problem. Here, we employ two deep learning methods based on Convolutional Neural Networks (CNNs). The first approach utilizes the Mask R-CNN algorithm to perform object detection on the major otolith reading axes. The second approach employs the U-Net architecture to perform semantic segmentation on the otolith image in order to segregate the regions of interest. For both methods, we applied a simple postprocessing to count the rings on the output masks returned, which corresponds to the age prediction. Multiple benchmark tests indicate the promising performance of our implemented approaches, comparable to recently published methods based on classical image processing and traditional CNN implementation. Furthermore, our algorithms showed higher robustness compared to the existing methods, while also having the capacity to extrapolate missing age groups and to adapt to a new domain or data source.
Cayetano, Arjay; Stransky, Christoph; Birk, Andreas; Brey, Thomas
An interactive AI-driven platform for fish age reading Journal Article
In: PLOS ONE, vol. 19, no. 11, pp. e0313934, 2024.
@article{FishAgeReading-PLOSONE24,
title = {An interactive AI-driven platform for fish age reading},
author = {Arjay Cayetano and Christoph Stransky and Andreas Birk and Thomas Brey},
url = {https://doi.org/10.1371/journal.pone.0313934},
doi = {10.1371/journal.pone.0313934},
year = {2024},
date = {2024-01-01},
journal = {PLOS ONE},
volume = {19},
number = {11},
pages = {e0313934},
abstract = {Fish age is an important biological variable required as part of routine stock assessment and analysis of fish population dynamics. Age estimates are traditionally obtained by human experts from the count of ring-like patterns along calcified structures such as otoliths. To automate the process and minimize human bias, modern methods have been designed utilizing the advances in the field of artificial intelligence (AI). While many AI-based methods have been shown to attain satisfactory accuracy, there are concerns regarding the lack of explainability of some early implementations. Consequently, new explainable AI-based approaches based on U-Net and Mask R-CNN have been recently published having direct compatibility with traditional ring counting procedures. Here we further extend this endeavor by creating an interactive website housing these explainable AI methods allowing age readers to be directly involved in the AI training and development. An important aspect of the platform presented in this article is that it allows the additional use of different advanced concepts of Machine Learning (ML) such as transfer learning, ensemble learning and continual learning, which are all shown to be effective in this study.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Fish age is an important biological variable required as part of routine stock assessment and analysis of fish population dynamics. Age estimates are traditionally obtained by human experts from the count of ring-like patterns along calcified structures such as otoliths. To automate the process and minimize human bias, modern methods have been designed utilizing the advances in the field of artificial intelligence (AI). While many AI-based methods have been shown to attain satisfactory accuracy, there are concerns regarding the lack of explainability of some early implementations. Consequently, new explainable AI-based approaches based on U-Net and Mask R-CNN have been recently published having direct compatibility with traditional ring counting procedures. Here we further extend this endeavor by creating an interactive website housing these explainable AI methods allowing age readers to be directly involved in the AI training and development. An important aspect of the platform presented in this article is that it allows the additional use of different advanced concepts of Machine Learning (ML) such as transfer learning, ensemble learning and continual learning, which are all shown to be effective in this study.
Hansen, Tim; Belenis, Bogdan; Firvida, Miguel Bande; Creutz, Tom; Birk, Andreas
Scanning Sonar Data from an Underwater Robot with Ground Truth Localization Journal Article
In: IEEE Access, vol. 12, pp. 129202-129211, 2024.
@article{ScanningSonar-GroundTruth-DFKIpool-Access24,
title = {Scanning Sonar Data from an Underwater Robot with Ground Truth Localization},
author = {Tim Hansen and Bogdan Belenis and Miguel Bande Firvida and Tom Creutz and Andreas Birk},
url = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10577110},
doi = {10.1109/ACCESS.2024.3420766},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
journal = {IEEE Access},
volume = {12},
pages = {129202-129211},
abstract = {Mechanical Scanning Sonars (MSS) are popular underwater sensors for Unmanned Underwater Vehicles (UUV) due to their low cost, small size, and low power consumption. But due to their simplicity, there are also many research challenges related to their usage. Unfortunately, there is also a lack of data with ground truth UUV localization. We provide MSS datasets using a UUV with standard navigation sensors, i.e., an Inertial Measurement Unit (IMU) and a Doppler Velocity Log (DVL). The UUV is globally localized with a high precision optical tracking system in a large research pool to provide ground truth. The data is of interest for multiple research areas related to MSS, e.g., extraction of range information, registration of sonar scans, and especially mapping including Simultaneous Localization and Mapping (SLAM). Different parameter settings and environment conditions are covered, e.g., dynamics in the scene. The IMU and DVL data is also of interest for research on navigation independent of the MSS data. Results from navigation and mapping with an Extended Kalman Filter (EKF) are in addition provided as baseline solutions.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mechanical Scanning Sonars (MSS) are popular underwater sensors for Unmanned Underwater Vehicles (UUV) due to their low cost, small size, and low power consumption. But due to their simplicity, there are also many research challenges related to their usage. Unfortunately, there is also a lack of data with ground truth UUV localization. We provide MSS datasets using a UUV with standard navigation sensors, i.e., an Inertial Measurement Unit (IMU) and a Doppler Velocity Log (DVL). The UUV is globally localized with a high precision optical tracking system in a large research pool to provide ground truth. The data is of interest for multiple research areas related to MSS, e.g., extraction of range information, registration of sonar scans, and especially mapping including Simultaneous Localization and Mapping (SLAM). Different parameter settings and environment conditions are covered, e.g., dynamics in the scene. The IMU and DVL data is also of interest for research on navigation independent of the MSS data. Results from navigation and mapping with an Extended Kalman Filter (EKF) are in addition provided as baseline solutions.
Hansen, Tim; Birk, Andreas
An Open-Source Solution for Fast and Accurate Underwater Mapping with a Low-Cost Mechanical Scanning Sonar Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2024.
@inproceedings{LowCostUnderwaterMapping-ICRA24,
title = {An Open-Source Solution for Fast and Accurate Underwater Mapping with a Low-Cost Mechanical Scanning Sonar},
author = {Tim Hansen and Andreas Birk},
url = {https://www.researchgate.net/publication/382989302_An_Open-Source_Solution_for_Fast_and_Accurate_Underwater_Mapping_with_a_Low-Cost_Mechanical_Scanning_Sonar},
doi = {10.1109/ICRA57147.2024.10609976},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {An open-source software framework is presented that allows real-time underwater mapping with popular marine robotics components, namely a BlueRobotics BlueROV2 with its standard Ping360 Mechanical Scanning Sonar (MSS) and A50 Doppler Velocity Log (DVL), which are low-cost devices for their respective types - if not even the most affordable ones on the market. The software runs with low computational power on a Raspberry Pi4. The framework builds upon Synthetic Scan Formation (SSF) where single MSS beams or scan-lines are embedded into a pose-graph. The rendering of scans is not as usual based on navigation, but it can improve in the spirit of a synthetic aperture. Scans formed from scan-lines can be optimized by online Simultaneous Localization and Mapping (SLAM) and improved scans in turn lead to improved registration results in subsequent steps. To this end, a combination of two different types of loop-closures is presented here. The framework is validated in three different test-environments, namely a pool, a test-tank with a gantry for ground truth motion, and the flooded basement of a WW-II submarine bunker. Among others, it is shown that there is an increased accuracy compared to conventional SLAM and that the software is usable in real-time during a mission with the low-cost hardware.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An open-source software framework is presented that allows real-time underwater mapping with popular marine robotics components, namely a BlueRobotics BlueROV2 with its standard Ping360 Mechanical Scanning Sonar (MSS) and A50 Doppler Velocity Log (DVL), which are low-cost devices for their respective types - if not even the most affordable ones on the market. The software runs with low computational power on a Raspberry Pi4. The framework builds upon Synthetic Scan Formation (SSF) where single MSS beams or scan-lines are embedded into a pose-graph. The rendering of scans is not as usual based on navigation, but it can improve in the spirit of a synthetic aperture. Scans formed from scan-lines can be optimized by online Simultaneous Localization and Mapping (SLAM) and improved scans in turn lead to improved registration results in subsequent steps. To this end, a combination of two different types of loop-closures is presented here. The framework is validated in three different test-environments, namely a pool, a test-tank with a gantry for ground truth motion, and the flooded basement of a WW-II submarine bunker. Among others, it is shown that there is an increased accuracy compared to conventional SLAM and that the software is usable in real-time during a mission with the low-cost hardware.
Birk, Andreas
SmileyNet - Towards the Prediction of the Lottery by Reading Tea Leaves with AI Miscellaneous
arXiv, 2024.
@misc{birk2024smileynetpredictionlottery,
title = {SmileyNet - Towards the Prediction of the Lottery by Reading Tea Leaves with AI},
author = {Andreas Birk},
url = {https://arxiv.org/abs/2407.21385},
doi = {10.48550/arXiv.2407.21385},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
number = {2407.21385},
publisher = {arXiv},
abstract = {We introduce SmileyNet, a novel neural network with psychic abilities. It is inspired by the fact that a positive mood can lead to improved cognitive capabilities including classification tasks. The network is hence presented in a first phase with smileys and an encouraging loss function is defined to bias it into a good mood. SmileyNet is then used to forecast the flipping of a coin based on an established method of Tasseology, namely by reading tea leaves. Training and testing in this second phase are done with a high-fidelity simulation based on real-world pixels sampled from a professional tea-reading cup. SmileyNet has an amazing accuracy of 72% to correctly predict the flip of a coin. Resnet-34, respectively YOLOv5 achieve only 49%, respectively 53%. It is then shown how multiple SmileyNets can be combined to win the lottery.},
howpublished = {arXiv},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
We introduce SmileyNet, a novel neural network with psychic abilities. It is inspired by the fact that a positive mood can lead to improved cognitive capabilities including classification tasks. The network is hence presented in a first phase with smileys and an encouraging loss function is defined to bias it into a good mood. SmileyNet is then used to forecast the flipping of a coin based on an established method of Tasseology, namely by reading tea leaves. Training and testing in this second phase are done with a high-fidelity simulation based on real-world pixels sampled from a professional tea-reading cup. SmileyNet has an amazing accuracy of 72% to correctly predict the flip of a coin. Resnet-34, respectively YOLOv5 achieve only 49%, respectively 53%. It is then shown how multiple SmileyNets can be combined to win the lottery.
2023
Birk, Andreas; Buda, Frederike; Hansen, Tim
Dating Wall Constructions from Brick Sizes in the Flooded Basement of a WW-II Submarine Bunker for the Digitization of Cultural Heritage Proceedings Article
In: MTS/IEEE OCEANS, 2023.
@inproceedings{Valentin-Basement-BrickSizes-Oceans23,
title = {Dating Wall Constructions from Brick Sizes in the Flooded Basement of a WW-II Submarine Bunker for the Digitization of Cultural Heritage},
author = {Andreas Birk and Frederike Buda and Tim Hansen},
url = {https://www.researchgate.net/publication/373885094_Underwater_Exploration_with_Sonar_of_the_Flooded_Basement_of_a_WW-II_Submarine_Bunker_in_the_Context_of_the_Digitization_of_Cultural_Heritage},
doi = {10.1109/OCEANSLimerick52467.2023.10244268},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {MTS/IEEE OCEANS},
abstract = {Results from the exploration of the flooded basement of the U-Boot Bunker Valentin are presented. The work is done in the context of the digitization of the bunker, which was constructed during World-War II using substantial amounts of forced labor and which is now a memorial. The presented research deals with walls in the basement for which it is of interest when they were constructed. Using a simple but efficient structured light system, the sizes of the bricks were measured. In combination with the changes in German standards for bricks after WW-II, this gives important clues about the time of construction of the investigated walls.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Results from the exploration of the flooded basement of the U-Boot Bunker Valentin are presented. The work is done in the context of the digitization of the bunker, which was constructed during World-War II using substantial amounts of forced labor and which is now a memorial. The presented research deals with walls in the basement for which it is of interest when they were constructed. Using a simple but efficient structured light system, the sizes of the bricks were measured. In combination with the changes in German standards for bricks after WW-II, this gives important clues about the time of construction of the investigated walls.
Hansen, Tim; Birk, Andreas
Using Registration with Fourier-SOFT in 2D (FS2D) for Robust Scan Matching of Sonar Range Data Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), 2023.
@inproceedings{FS2D-MSSregistration-ICRA23,
title = {Using Registration with Fourier-SOFT in 2D (FS2D) for Robust Scan Matching of Sonar Range Data},
author = {Tim Hansen and Andreas Birk},
url = {https://www.researchgate.net/publication/372130798_Using_Registration_with_Fourier-SOFT_in_2D_FS2D_for_Robust_Scan_Matching_of_Sonar_Range_Data},
doi = {10.1109/ICRA48891.2023.10160519},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
abstract = {In this paper, we introduce Fourier-SOFT 2D (FS2D) as a new robust registration method. FS2D operates in the frequency domain where it exploits the well-known decoupling of rotation and translation. The challenging part of determining the rotation parameter is solved here based on a projection of the Fourier magnitude on a sphere and the SO(3) Fourier Transform (SOFT). The underlying use case is underwater mapping with sonar, i.e., with very noisy and partially overlapping environment data under non-trivial localization and navigation challenges. Fourier-SOFT 2D is compared with openly available registration methods on two real-world datasets and a simulated dataset. Results show the robustness of FS2D, i.e., its capabilities to handle large amounts of noise and occlusions of consecutive scans. The implementation in C++ is openly available.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, we introduce Fourier-SOFT 2D (FS2D) as a new robust registration method. FS2D operates in the frequency domain where it exploits the well-known decoupling of rotation and translation. The challenging part of determining the rotation parameter is solved here based on a projection of the Fourier magnitude on a sphere and the SO(3) Fourier Transform (SOFT). The underlying use case is underwater mapping with sonar, i.e., with very noisy and partially overlapping environment data under non-trivial localization and navigation challenges. Fourier-SOFT 2D is compared with openly available registration methods on two real-world datasets and a simulated dataset. Results show the robustness of FS2D, i.e., its capabilities to handle large amounts of noise and occlusions of consecutive scans. The implementation in C++ is openly available.
Hansen, Tim; Buda, Frederike; Birk, Andreas
Underwater Exploration with Sonar of the Flooded Basement of a WW-II Submarine Bunker in the Context of Digitization of Cultural Heritage Proceedings Article
In: MTS/IEEE OCEANS, pp. 1-5, 2023.
@inproceedings{Valentin-BasementMapping-Oceans23,
title = {Underwater Exploration with Sonar of the Flooded Basement of a WW-II Submarine Bunker in the Context of Digitization of Cultural Heritage},
author = {Tim Hansen and Frederike Buda and Andreas Birk},
url = {https://www.researchgate.net/publication/373885094_Underwater_Exploration_with_Sonar_of_the_Flooded_Basement_of_a_WW-II_Submarine_Bunker_in_the_Context_of_the_Digitization_of_Cultural_Heritage},
doi = {10.1109/OCEANSLimerick52467.2023.10244377},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
booktitle = {MTS/IEEE OCEANS},
pages = {1-5},
abstract = {Underwater mapping based on the registration of data from a mechanical scanning sonar (MSS) is presented. The specific use-case is the exploration of the flooded basement of the U-Boot Bunker Valentin in the context of the digitization of this World-War-II memorial. The visibility in the basement, which is inaccessible for divers due to safety reasons, is very poor. Hence, maps are very helpful if not essential for the exploration of this environment, for which - despite its rather recent history - open questions exist. The mapping is conducted with a very robust method in form of Fourier-SOFT 2D (FS2D), which is well suited for real-time registration of the very noisy scans generated with simple motion-compensation from the MSS data. More precisely, the exploration of the bunker basement is conducted with low-cost hardware in form of a BlueROV2 vehicle, simple navigation sensors, and a Ping360 MSS. We show that reasonable maps can be generated in real-time with this equipment when using FS2D.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater mapping based on the registration of data from a mechanical scanning sonar (MSS) is presented. The specific use-case is the exploration of the flooded basement of the U-Boot Bunker Valentin in the context of the digitization of this World-War-II memorial. The visibility in the basement, which is inaccessible for divers due to safety reasons, is very poor. Hence, maps are very helpful if not essential for the exploration of this environment, for which - despite its rather recent history - open questions exist. The mapping is conducted with a very robust method in form of Fourier-SOFT 2D (FS2D), which is well suited for real-time registration of the very noisy scans generated with simple motion-compensation from the MSS data. More precisely, the exploration of the bunker basement is conducted with low-cost hardware in form of a BlueROV2 vehicle, simple navigation sensors, and a Ping360 MSS. We show that reasonable maps can be generated in real-time with this equipment when using FS2D.
Hansen, T.; Birk, A.
Synthetic Scan Formation for Underwater Mapping with Low-Cost Mechanical Scanning Sonars (MSS) Journal Article
In: IEEE Access, vol. 11, pp. 96,854-96,864, 2023, ISSN: 2169-3536.
@article{SyntheticScanFormation-SSF-Access23,
title = {Synthetic Scan Formation for Underwater Mapping with Low-Cost Mechanical Scanning Sonars (MSS)},
author = {T. Hansen and A. Birk},
url = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10239384},
doi = {10.1109/access.2023.3312186},
issn = {2169-3536},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {IEEE Access},
volume = {11},
pages = {96,854-96,864},
abstract = {Mechanical Scanning Sonars (MSS) are popular devices for Unmanned Underwater Vehicles (UUV), i.e., Remotely Operated Vehicles (ROV) and Autonomous Underwater Vehicles (AUV), as they function under low visibility conditions and over extended ranges. They are comparatively low-cost and easy to integrate. But they require motion-compensation due to the low updates rates caused by the mechanical scanning and the slow speed of sound. We present here a new form of scan formation for MSS where the data from single beams is embedded into a pose-graph. The rendering of scans is not as usual based on only core navigation sensors, but it can improve in the spirit of a synthetic aperture. To this end, online Simultaneous Localization and Mapping (SLAM) is used to form scans from the single beams. These can be optimized and improved scans in turn lead to improved registration results in subsequent steps. This Synthetic Scan Formation (SSF) leads to better mapping results than state-of-the-art SLAM with MSS. The method is validated with several real-world experiments. First, different trajectories with precise ground-truth in a pool with a gantry set-up are used. Second, results from field trials in a WW-II submarine bunker are presented. It is shown that there are clear quantitative and qualitative improvements, and that SSF can be used in real-time for mapping during a mission.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mechanical Scanning Sonars (MSS) are popular devices for Unmanned Underwater Vehicles (UUV), i.e., Remotely Operated Vehicles (ROV) and Autonomous Underwater Vehicles (AUV), as they function under low visibility conditions and over extended ranges. They are comparatively low-cost and easy to integrate. But they require motion-compensation due to the low updates rates caused by the mechanical scanning and the slow speed of sound. We present here a new form of scan formation for MSS where the data from single beams is embedded into a pose-graph. The rendering of scans is not as usual based on only core navigation sensors, but it can improve in the spirit of a synthetic aperture. To this end, online Simultaneous Localization and Mapping (SLAM) is used to form scans from the single beams. These can be optimized and improved scans in turn lead to improved registration results in subsequent steps. This Synthetic Scan Formation (SSF) leads to better mapping results than state-of-the-art SLAM with MSS. The method is validated with several real-world experiments. First, different trajectories with precise ground-truth in a pool with a gantry set-up are used. Second, results from field trials in a WW-II submarine bunker are presented. It is shown that there are clear quantitative and qualitative improvements, and that SSF can be used in real-time for mapping during a mission.
2022
Birk, Andreas; Buda, Frederike; Bülow, Heiko; Chavez, Arturo Gomez; Müller, Christian A.; Timpe, Julia
Digitizing a Gigantic Nazi Construction: 3D-Mapping of Bunker Valentin in Bremen Book Chapter
In: Julia, Timpe; Frederike, Buda (Ed.): Writing the Digital History of Nazi Germany, pp. 133-168, De Gruyter, Berlin, Boston, 2022.
@inbook{CulturalHeritage-Valentin3D-DeGruyter22,
title = {Digitizing a Gigantic Nazi Construction: 3D-Mapping of Bunker Valentin in Bremen},
author = {Andreas Birk and Frederike Buda and Heiko Bülow and Arturo Gomez Chavez and Christian A. Müller and Julia Timpe},
editor = {Timpe Julia and Buda Frederike},
url = {https://doi.org/10.1515/9783110714692-006},
doi = {doi:10.1515/9783110714692-006},
year = {2022},
date = {2022-01-01},
booktitle = {Writing the Digital History of Nazi Germany},
pages = {133-168},
publisher = {De Gruyter},
address = {Berlin, Boston},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas
A Survey of Underwater Human-Robot Interaction (U-HRI) Journal Article
In: Current Robotics Reports, Springer Nature, vol. 3, pp. 199-211, 2022.
@article{UHRI-Survey-CRR22,
title = {A Survey of Underwater Human-Robot Interaction (U-HRI)},
author = {Andreas Birk},
url = {https://link.springer.com/article/10.1007/s43154-022-00092-7},
doi = {10.1007/s43154-022-00092-7},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
journal = {Current Robotics Reports, Springer Nature},
volume = {3},
pages = {199-211},
abstract = {This review provides an overview of the current state of the art in Underwater Human-Robot Interaction (U-HRI), which is an area that is quite different from standard Human-Robot Interaction (HRI). This is due to several reasons. First of all, there are the particular properties of water as a medium, e.g., the strong attenuation of radio-frequency (RF) signals or the physics of underwater image formation. Second, divers are bound to special equipment, e.g., the breathing apparatus, which makes, for example, speech recognition challenging, if not impossible. Third, typical collaborative marine missions primarily requires a high amount of communication from the diver to the robot, which accordingly receives a lot of attention in U-HRI research.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This review provides an overview of the current state of the art in Underwater Human-Robot Interaction (U-HRI), which is an area that is quite different from standard Human-Robot Interaction (HRI). This is due to several reasons. First of all, there are the particular properties of water as a medium, e.g., the strong attenuation of radio-frequency (RF) signals or the physics of underwater image formation. Second, divers are bound to special equipment, e.g., the breathing apparatus, which makes, for example, speech recognition challenging, if not impossible. Third, typical collaborative marine missions primarily requires a high amount of communication from the diver to the robot, which accordingly receives a lot of attention in U-HRI research.
2021
Birk, Andreas
In: Science Robotics, vol. 6, no. 55, 2021.
@article{OAS-SchedlEtal-Commentary-ScienceRob21,
title = {Seeing through the forest and the trees with drones: Signal-processing of thermal images that are autonomously collected by a drone detects people in densely occluded forests},
author = {Andreas Birk},
url = {https://robotics.sciencemag.org/content/6/55/eabj3947},
year = {2021},
date = {2021-01-01},
journal = {Science Robotics},
volume = {6},
number = {55},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Birk, Andreas; Dineva, Evelina
Improved Students’ Performance in an Online Robotics Class During COVID-19: Do only Strong Students Profit? Proceedings Article
In: Merdan, Munir; Lepuschitz, Wilfried; Koppensteiner, Gottfried; Balogh, Richard; Obdrzalek, David (Ed.): Robotics in Education (RiE), Advances in Intelligent Systems and Computing, pp. 134-145, Springer International Publishing, 2021, ISBN: 978-3-030-82544-7.
@inproceedings{Edu-RoboticsCourseOnline-DoOnlyStrongStudentsProfit-RiE21,
title = {Improved Students’ Performance in an Online Robotics Class During COVID-19: Do only Strong Students Profit?},
author = {Andreas Birk and Evelina Dineva},
editor = {Munir Merdan and Wilfried Lepuschitz and Gottfried Koppensteiner and Richard Balogh and David Obdrzalek},
url = {https://www.researchgate.net/publication/353631426_Improved_Students'_Performance_in_an_Online_Robotics_Class_During_COVID-19_Do_only_Strong_Students_Profit},
doi = {10.1007/978-3-030-82544-7_13},
isbn = {978-3-030-82544-7},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
booktitle = {Robotics in Education (RiE), Advances in Intelligent Systems and Computing},
volume = {1359},
pages = {134-145},
publisher = {Springer International Publishing},
abstract = {Observations from the online teaching of a robotics class during the CoViD-19 pandemic caused by SARS-CoV-2 also known as the Corona virus are presented. The online teaching of the course lead to an unexpected increase in the grade performance of the students – despite the fact that the exams were at the same level of difficulty as in the years before. There is some evidence that the asynchronous online teaching catered for some of the general interests and learning trends in generation-Z students. This paper addresses the question whether the general study performance of a student is a possible factor in whether the student profited more or less from the online version, i.e., whether there is a correlation between the improved course grade with the Grade Performance Average (GPA) of students.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Observations from the online teaching of a robotics class during the CoViD-19 pandemic caused by SARS-CoV-2 also known as the Corona virus are presented. The online teaching of the course lead to an unexpected increase in the grade performance of the students – despite the fact that the exams were at the same level of difficulty as in the years before. There is some evidence that the asynchronous online teaching catered for some of the general interests and learning trends in generation-Z students. This paper addresses the question whether the general study performance of a student is a possible factor in whether the student profited more or less from the online version, i.e., whether there is a correlation between the improved course grade with the Grade Performance Average (GPA) of students.
Birk, Andreas; Dineva, Evelina; Maurelli, Francesco; Nabor, Andreas
A Robotics Course during COVID-19: Lessons Learned and Best Practices for Online Teaching beyond the Pandemic Journal Article
In: Robotics, vol. 10, no. 5, 2021, ISSN: 2218-6581.
@article{EDU-MDPI20-RoboticsCourseCoViD,
title = {A Robotics Course during COVID-19: Lessons Learned and Best Practices for Online Teaching beyond the Pandemic},
author = {Andreas Birk and Evelina Dineva and Francesco Maurelli and Andreas Nabor},
url = {https://www.mdpi.com/2218-6581/10/1/5},
doi = {10.3390/robotics10010005},
issn = {2218-6581},
year = {2021},
date = {2021-01-01},
journal = {Robotics},
volume = {10},
number = {5},
abstract = {The article describes observations from the online teaching of a robotics class during the COVID-19 pandemic caused by SARS-CoV-2, also known as the coronavirus. The changes in the course structure and in the provided material lead to an unexpected increase in the grade performance of the students. The article provides a description and an analysis of the effects and their possible causes. In addition to a grade-performance analysis, further data from a university-wide and from a course-specific survey are used. The analysis of the effects and their possible causes is furthermore discussed in relation to the educational research literature. Some evidence for the general findings is provided, which are of interest for online teaching or blended learning in general, respectively, for teaching in robotics and related areas. These include some evidence for the benefits of asynchronous online teaching and for the role of social interaction, which may happen in self-organized, smaller peer groups, even without the intervention of the instructor. The findings and the extensive pointers to the literature can also provide useful guidelines for instructors of robotics courses when considering the use of online or blended teaching in the future beyond the COVID-19 pandemic.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The article describes observations from the online teaching of a robotics class during the COVID-19 pandemic caused by SARS-CoV-2, also known as the coronavirus. The changes in the course structure and in the provided material lead to an unexpected increase in the grade performance of the students. The article provides a description and an analysis of the effects and their possible causes. In addition to a grade-performance analysis, further data from a university-wide and from a course-specific survey are used. The analysis of the effects and their possible causes is furthermore discussed in relation to the educational research literature. Some evidence for the general findings is provided, which are of interest for online teaching or blended learning in general, respectively, for teaching in robotics and related areas. These include some evidence for the benefits of asynchronous online teaching and for the role of social interaction, which may happen in self-organized, smaller peer groups, even without the intervention of the instructor. The findings and the extensive pointers to the literature can also provide useful guidelines for instructors of robotics courses when considering the use of online or blended teaching in the future beyond the COVID-19 pandemic.
Birk, Andreas; Simunovic, Dora
Robotics Labs and Other Hands-On Teaching During COVID-19: Change Is Here to Stay? Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 28, no. 4, 2021.
@article{Edu-OnlineRoboticsLabs-RAM21,
title = {Robotics Labs and Other Hands-On Teaching During COVID-19: Change Is Here to Stay?},
author = {Andreas Birk and Dora Simunovic},
url = {https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9523602},
doi = {10.1109/MRA.2021.3102979},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {28},
number = {4},
abstract = {The Covid-19 pandemic had profound effects on almost all aspects of daily life including education where it lead to a rapid and massive switch to online teaching. There is evidence that the responses to the pandemic will have lasting effects in higher education and that it has accelerated the trend that (elements of) online educations will be the new normal in higher education. Students expect for example as a main change for universities in ten years time that most lectures will be online.
But there are disciplines like educational robotics, where laboratory work, respectively a hands-on experience is considered to be essential for education. As pointed out in the literature, almost all work in educational robotics refers in one form or the other to Papert's theory of constructionism, stressing the benefits of "learning-by-making". This importance is also highlighted by more recent educational theories like Problem Based Learning as well as Project Based Learning, especially when used in the context of STEM (Science, Technology, Engineering, and Math) and in particular engineering education. An obvious questions is hence how Covid-19 affected robotics teaching where physical presence in labs is often considered to be sacrosanct for hands-on experience, e.g., by requiring engineering students in even distance education programs to come to campus for in-presence lab sessions.
In this article, we provide the results of an international survey on the effects of Covid-19 on robotics laboratory courses and other forms of hands-on education in tertiary robotics education. The goal of this study is to provide an overview of the way instructors handled the challenges of Covid-19 in teaching hands-on robotics courses, and to identify possible effects and trends that are of relevance also in the future beyond the pandemic. As an important finding, the study provides indications that physical access to equipment in laboratories is on the one hand considered to be essential, while on the other hand, the alternative forms of hands-on teaching during the pandemic lead to increased grade performances and the instructors report plans to make the teaching changes permanent. The effects of the pandemic hence indicate that there may be substantial room for hands-on education in robotics without the requirement for classical labs with physical access to robotics equipment in on-campus facilities.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The Covid-19 pandemic had profound effects on almost all aspects of daily life including education where it lead to a rapid and massive switch to online teaching. There is evidence that the responses to the pandemic will have lasting effects in higher education and that it has accelerated the trend that (elements of) online educations will be the new normal in higher education. Students expect for example as a main change for universities in ten years time that most lectures will be online.
But there are disciplines like educational robotics, where laboratory work, respectively a hands-on experience is considered to be essential for education. As pointed out in the literature, almost all work in educational robotics refers in one form or the other to Papert's theory of constructionism, stressing the benefits of "learning-by-making". This importance is also highlighted by more recent educational theories like Problem Based Learning as well as Project Based Learning, especially when used in the context of STEM (Science, Technology, Engineering, and Math) and in particular engineering education. An obvious questions is hence how Covid-19 affected robotics teaching where physical presence in labs is often considered to be sacrosanct for hands-on experience, e.g., by requiring engineering students in even distance education programs to come to campus for in-presence lab sessions.
In this article, we provide the results of an international survey on the effects of Covid-19 on robotics laboratory courses and other forms of hands-on education in tertiary robotics education. The goal of this study is to provide an overview of the way instructors handled the challenges of Covid-19 in teaching hands-on robotics courses, and to identify possible effects and trends that are of relevance also in the future beyond the pandemic. As an important finding, the study provides indications that physical access to equipment in laboratories is on the one hand considered to be essential, while on the other hand, the alternative forms of hands-on teaching during the pandemic lead to increased grade performances and the instructors report plans to make the teaching changes permanent. The effects of the pandemic hence indicate that there may be substantial room for hands-on education in robotics without the requirement for classical labs with physical access to robotics equipment in on-campus facilities.
But there are disciplines like educational robotics, where laboratory work, respectively a hands-on experience is considered to be essential for education. As pointed out in the literature, almost all work in educational robotics refers in one form or the other to Papert's theory of constructionism, stressing the benefits of "learning-by-making". This importance is also highlighted by more recent educational theories like Problem Based Learning as well as Project Based Learning, especially when used in the context of STEM (Science, Technology, Engineering, and Math) and in particular engineering education. An obvious questions is hence how Covid-19 affected robotics teaching where physical presence in labs is often considered to be sacrosanct for hands-on experience, e.g., by requiring engineering students in even distance education programs to come to campus for in-presence lab sessions.
In this article, we provide the results of an international survey on the effects of Covid-19 on robotics laboratory courses and other forms of hands-on education in tertiary robotics education. The goal of this study is to provide an overview of the way instructors handled the challenges of Covid-19 in teaching hands-on robotics courses, and to identify possible effects and trends that are of relevance also in the future beyond the pandemic. As an important finding, the study provides indications that physical access to equipment in laboratories is on the one hand considered to be essential, while on the other hand, the alternative forms of hands-on teaching during the pandemic lead to increased grade performances and the instructors report plans to make the teaching changes permanent. The effects of the pandemic hence indicate that there may be substantial room for hands-on education in robotics without the requirement for classical labs with physical access to robotics equipment in on-campus facilities.
Bülow, Heiko; Birk, Andreas
Registration of Magnetic Resonance Tomography (MRT) Data with a Low Frequency Adaption of Fourier-Mellin-SOFT (LF-FMS) Journal Article
In: Sensors, vol. 21, no. 8, 2021.
@article{LF-FMS-Medical-Sensors21,
title = {Registration of Magnetic Resonance Tomography (MRT) Data with a Low Frequency Adaption of Fourier-Mellin-SOFT (LF-FMS)},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.mdpi.com/1424-8220/21/8/2581},
doi = {10.3390/s21082581},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
journal = {Sensors},
volume = {21},
number = {8},
abstract = {Fourier-Mellin-SOFT (FMS) is a rigid 3D registration method, which allows the robust registration of 3 degrees-of-freedom (dof) rotation, 1-dof scale, and 3-dof translation between scans on discrete grids. FMS is based on a spectral decomposition of these 7-dof. This complete spectral representation of the input data enables an adaption to certain frequency ranges. This special property is used here to focus on relevant mutual 3D information between bone structures with a Low Frequency adaptation of FMS (LF-FMS), that is, it is utilized for matching and concurrently determining corresponding transformation parameters. This process is applied on a set of Magnetic Resonance Tomography (MRT) data representing the hand region, in particular the carpal bone area, in a sequence of different hand positions. This data set is available for different probands, which allows a comparison of resulting parameter plots and furthermore matching in between bone structures.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Fourier-Mellin-SOFT (FMS) is a rigid 3D registration method, which allows the robust registration of 3 degrees-of-freedom (dof) rotation, 1-dof scale, and 3-dof translation between scans on discrete grids. FMS is based on a spectral decomposition of these 7-dof. This complete spectral representation of the input data enables an adaption to certain frequency ranges. This special property is used here to focus on relevant mutual 3D information between bone structures with a Low Frequency adaptation of FMS (LF-FMS), that is, it is utilized for matching and concurrently determining corresponding transformation parameters. This process is applied on a set of Magnetic Resonance Tomography (MRT) data representing the hand region, in particular the carpal bone area, in a sequence of different hand positions. This data set is available for different probands, which allows a comparison of resulting parameter plots and furthermore matching in between bone structures.
Chavez, Arturo Gomez; Ranieri, Andrea; Chiarella, Davide; Birk, Andreas
Underwater Vision-Based Gesture Recognition: A Robustness Validation for Safe Human-Robot Interaction Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 28, no. 3, pp. 67-78, 2021.
@article{UnderwaterHMI-CADDY-RAM21,
title = {Underwater Vision-Based Gesture Recognition: A Robustness Validation for Safe Human-Robot Interaction},
author = {Arturo Gomez Chavez and Andrea Ranieri and Davide Chiarella and Andreas Birk},
url = {https://www.researchgate.net/publication/351757730_Underwater_Vision-Based_Gesture_Recognition_A_Robustness_Validation_for_Safe_Human-Robot_Interaction},
doi = {10.1109/MRA.2021.3075560},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {28},
number = {3},
pages = {67-78},
abstract = {Underwater robotics requires very reliable and safe operations. This holds especially true for missions in cooperation with divers who are-despite the significant advancements of marine robotics in recent years-still essential for many underwater operations. Possible application cases of underwater human-robot collaboration include marine science, archeology, oil and gas production, handling of unexploded ordnance (e.g., from World War II ammunition dumped in the seas), or the inspection and maintenance of marine infrastructure like pipelines, harbors, or renewable energy installations, to name just a few examples.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Underwater robotics requires very reliable and safe operations. This holds especially true for missions in cooperation with divers who are-despite the significant advancements of marine robotics in recent years-still essential for many underwater operations. Possible application cases of underwater human-robot collaboration include marine science, archeology, oil and gas production, handling of unexploded ordnance (e.g., from World War II ammunition dumped in the seas), or the inspection and maintenance of marine infrastructure like pipelines, harbors, or renewable energy installations, to name just a few examples.
Maurelli, Francesco; Dineva, Evelina; Nabor, Andreas; Birk, Andreas
Robotics and Intelligent Systems: a new curriculum development and adaptations needed in Coronavirus times Proceedings Article
In: Merdan, Munir; Lepuschitz, Wilfried; Koppensteiner, Gottfried; Balogh, Richard; Obdrzalek, David (Ed.): Robotics in Education (RiE), Advances in Intelligent Systems and Computing, pp. 134-145, Springer International Publishing, 2021, ISBN: 978-3-030-82544-7.
@inproceedings{Edu-RIS-Corona-RiE21,
title = {Robotics and Intelligent Systems: a new curriculum development and adaptations needed in Coronavirus times},
author = {Francesco Maurelli and Evelina Dineva and Andreas Nabor and Andreas Birk},
editor = {Munir Merdan and Wilfried Lepuschitz and Gottfried Koppensteiner and Richard Balogh and David Obdrzalek},
url = {https://link.springer.com/chapter/10.1007/978-3-030-82544-7_9},
doi = {10.1007/978-3-030-82544-7_9},
isbn = {978-3-030-82544-7},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
booktitle = {Robotics in Education (RiE), Advances in Intelligent Systems and Computing},
volume = {1359},
pages = {134-145},
publisher = {Springer International Publishing},
abstract = {This paper presents the curriculum development of the Robotics and Intelligent Systems Bachelor of Science program at Jacobs University, one of the very few bachelor robotics-focused programs in Germany. After highlighting the quality management measures and feedback received on the older Intelligent Mobile Systems, the new curriculum is highlighted, together with implementation challenges linked to Covid19 pandemic. Student satisfaction plots provide an initial assessment of the developed program.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents the curriculum development of the Robotics and Intelligent Systems Bachelor of Science program at Jacobs University, one of the very few bachelor robotics-focused programs in Germany. After highlighting the quality management measures and feedback received on the older Intelligent Mobile Systems, the new curriculum is highlighted, together with implementation challenges linked to Covid19 pandemic. Student satisfaction plots provide an initial assessment of the developed program.
Thosar, Madhura; Mueller, Christian A.; Jäger, Georg; Schleiss, Johannes; Pulugu, Narender; Mallikarjun, Ravi; Jeevangekar, Sai Vivek Rao; Birk, Andreas; Pfingsthorn, Max; Zug, Sebastian
From a Multi-modal Property Dataset to Robot-centric Conceptual Knowledge About Household Objects Journal Article
In: Frontiers in Robotics and AI, vol. 8, 2021.
@article{RobocentricObjectPerception-Household-FRA21,
title = {From a Multi-modal Property Dataset to Robot-centric Conceptual Knowledge About Household Objects},
author = {Madhura Thosar and Christian A. Mueller and Georg Jäger and Johannes Schleiss and Narender Pulugu and Ravi Mallikarjun and Sai Vivek Rao Jeevangekar and Andreas Birk and Max Pfingsthorn and Sebastian Zug},
url = {https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2021.476084/full},
doi = {10.3389/frobt.2021.476084},
year = {2021},
date = {2021-01-01},
urldate = {2021-01-01},
journal = {Frontiers in Robotics and AI},
volume = {8},
abstract = {Conceptual knowledge about objects is essential for humans, as well as for animals, to interact with their environment. On this basis, the objects can be understood as tools, a selection process can be implemented and their usage can be planned in order to achieve a specific goal. The conceptual knowledge, in this case, is primarily concerned about the physical properties and functional properties observed in the objects. Similarly tool-use applications in robotics require such conceptual knowledge about objects for substitute selection among other purposes. State-of-the-art methods employ a top-down approach where hand-crafted symbolic knowledge, which is defined from a human perspective, is grounded into sensory data afterwards. However, due to different sensing and acting capabilities of robots, a robot's conceptual understanding of objects (e.g., light/heavy) will vary and therefore should be generated from the robot's perspective entirely, which entails robot-centric conceptual knowledge about objects. A similar bottom-up argument has been put forth in cognitive science that humans and animals alike develop conceptual understanding of objects based on their own perceptual experiences with objects. With this goal in mind, we propose an extensible property estimation framework which consists of estimations methods to obtain the quantitative measurements of physical properties (rigidity, weight, etc.) and functional properties (containment, support, etc.) from household objects. This property estimation forms the basis for our second contribution: Generation of robot-centric conceptual knowledge. Our approach employs unsupervised clustering methods to transform numerical property data into symbols, and Bivariate Joint Frequency Distributions and Sample Proportion to generate conceptual knowledge about objects using the robot-centric symbols. A preliminary implementation of the proposed framework is employed to acquire a dataset comprising six physical and four functional properties of 110 household objects. This Robot-Centric dataSet (RoCS) is used to evaluate the framework regarding the property estimation methods and the semantics of the considered properties within the dataset. Furthermore, the dataset includes the derived robot-centric conceptual knowledge using the proposed framework. The application of the conceptual knowledge about objects is then evaluated by examining its usefulness in a tool substitution scenario.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Conceptual knowledge about objects is essential for humans, as well as for animals, to interact with their environment. On this basis, the objects can be understood as tools, a selection process can be implemented and their usage can be planned in order to achieve a specific goal. The conceptual knowledge, in this case, is primarily concerned about the physical properties and functional properties observed in the objects. Similarly tool-use applications in robotics require such conceptual knowledge about objects for substitute selection among other purposes. State-of-the-art methods employ a top-down approach where hand-crafted symbolic knowledge, which is defined from a human perspective, is grounded into sensory data afterwards. However, due to different sensing and acting capabilities of robots, a robot's conceptual understanding of objects (e.g., light/heavy) will vary and therefore should be generated from the robot's perspective entirely, which entails robot-centric conceptual knowledge about objects. A similar bottom-up argument has been put forth in cognitive science that humans and animals alike develop conceptual understanding of objects based on their own perceptual experiences with objects. With this goal in mind, we propose an extensible property estimation framework which consists of estimations methods to obtain the quantitative measurements of physical properties (rigidity, weight, etc.) and functional properties (containment, support, etc.) from household objects. This property estimation forms the basis for our second contribution: Generation of robot-centric conceptual knowledge. Our approach employs unsupervised clustering methods to transform numerical property data into symbols, and Bivariate Joint Frequency Distributions and Sample Proportion to generate conceptual knowledge about objects using the robot-centric symbols. A preliminary implementation of the proposed framework is employed to acquire a dataset comprising six physical and four functional properties of 110 household objects. This Robot-Centric dataSet (RoCS) is used to evaluate the framework regarding the property estimation methods and the semantics of the considered properties within the dataset. Furthermore, the dataset includes the derived robot-centric conceptual knowledge using the proposed framework. The application of the conceptual knowledge about objects is then evaluated by examining its usefulness in a tool substitution scenario.
2020
Birk, Andreas; Dineva, Evelina; Maurelli, Francesco; Nabor, Andreas
A Robotics Course during CoViD-19: Lessons Learned and Best Practices for Online Teaching beyond the Pandemic Journal Article
In: Robotics, 2020.
@article{OnlineTeaching-RoboticsCourse-2020,
title = {A Robotics Course during CoViD-19: Lessons Learned and Best Practices for Online Teaching beyond the Pandemic},
author = {Andreas Birk and Evelina Dineva and Francesco Maurelli and Andreas Nabor},
url = {https://www.mdpi.com/2218-6581/10/1/5/pdf?version=1609753222},
doi = {10.3390/robotics10010005},
year = {2020},
date = {2020-01-01},
urldate = {2020-01-01},
journal = {Robotics},
abstract = {The article describes observations from the online teaching of a robotics class during the COVID-19 pandemic caused by SARS-CoV-2, also known as the coronavirus. The changes in the course structure and in the provided material lead to an unexpected increase in the grade performance of the students. The article provides a description and an analysis of the effects and their possible causes. In addition to a grade-performance analysis, further data from a university-wide and from a course-specific survey are used. The analysis of the effects and their possible causes is furthermore discussed in relation to the educational research literature. Some evidence for the general findings is provided, which are of interest for online teaching or blended learning in general, respectively, for teaching in robotics and related areas. These include some evidence for the benefits of asynchronous online teaching and for the role of social interaction, which may happen in self-organized, smaller peer groups, even without the intervention of the instructor. The findings and the extensive pointers to the literature can also provide useful guidelines for instructors of robotics courses when considering the use of online or blended teaching in the future beyond the COVID-19 pandemic.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The article describes observations from the online teaching of a robotics class during the COVID-19 pandemic caused by SARS-CoV-2, also known as the coronavirus. The changes in the course structure and in the provided material lead to an unexpected increase in the grade performance of the students. The article provides a description and an analysis of the effects and their possible causes. In addition to a grade-performance analysis, further data from a university-wide and from a course-specific survey are used. The analysis of the effects and their possible causes is furthermore discussed in relation to the educational research literature. Some evidence for the general findings is provided, which are of interest for online teaching or blended learning in general, respectively, for teaching in robotics and related areas. These include some evidence for the benefits of asynchronous online teaching and for the role of social interaction, which may happen in self-organized, smaller peer groups, even without the intervention of the instructor. The findings and the extensive pointers to the literature can also provide useful guidelines for instructors of robotics courses when considering the use of online or blended teaching in the future beyond the COVID-19 pandemic.
Bülow, Heiko; Birk, Andreas
Synthetic Aperture Sonar (SAS) without Navigation: Scan Registration as Basis for Near Field Synthetic Imaging in 2D Journal Article
In: Sensors, vol. 20, no. 16, 2020.
@article{USAS2D-Sensors20,
title = {Synthetic Aperture Sonar (SAS) without Navigation: Scan Registration as Basis for Near Field Synthetic Imaging in 2D},
author = {Heiko Bülow and Andreas Birk},
url = {https://doi.org/10.3390/s20164440},
year = {2020},
date = {2020-01-01},
urldate = {2020-01-01},
journal = {Sensors},
volume = {20},
number = {16},
abstract = {Sonars are essential for underwater sensing as they can operate over extended ranges and in poor visibility conditions. The use of a synthetic aperture is a popular approach to increase the resolution of sonars, i.e., the sonar with its N transducers is positioned at k places to generate a virtual sensor with 𝑘𝑁
transducers. The state of the art for synthetic aperture sonar (SAS) is strongly coupled to constraints, especially with respect to the trajectory of the placements and the need for good navigation data. In this article, we introduce an approach to SAS using registration of scans from single arrays, i.e., at individual poses of arbitrary trajectories, hence avoiding the need for navigation data of conventional SAS systems. The approach is introduced here for the near field using the coherent phase information of sonar scans. A Delay and Sum (D&S) beamformer (BF) is used, which directly operates on pixel/voxel form on a Cartesian grid supporting the registration. It is shown that this pixel/voxel-based registration and the coherent processing of several scans forming a synthetic aperture yields substantial improvements of the image resolution. The experimental evaluation is done with an advanced simulation tool generating realistic 2D sonar array data, i.e., with simulations of a linear 1D antenna reconstructing 2D images. For the image registration of the raw sonar scans, a robust implementation of a spectral method is presented. Furthermore, analyses with respect to the trajectories of the sensor locations are provided to remedy possible grating lobes due to the gaping positions of the transmitter devices.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Sonars are essential for underwater sensing as they can operate over extended ranges and in poor visibility conditions. The use of a synthetic aperture is a popular approach to increase the resolution of sonars, i.e., the sonar with its N transducers is positioned at k places to generate a virtual sensor with 𝑘𝑁
transducers. The state of the art for synthetic aperture sonar (SAS) is strongly coupled to constraints, especially with respect to the trajectory of the placements and the need for good navigation data. In this article, we introduce an approach to SAS using registration of scans from single arrays, i.e., at individual poses of arbitrary trajectories, hence avoiding the need for navigation data of conventional SAS systems. The approach is introduced here for the near field using the coherent phase information of sonar scans. A Delay and Sum (D&S) beamformer (BF) is used, which directly operates on pixel/voxel form on a Cartesian grid supporting the registration. It is shown that this pixel/voxel-based registration and the coherent processing of several scans forming a synthetic aperture yields substantial improvements of the image resolution. The experimental evaluation is done with an advanced simulation tool generating realistic 2D sonar array data, i.e., with simulations of a linear 1D antenna reconstructing 2D images. For the image registration of the raw sonar scans, a robust implementation of a spectral method is presented. Furthermore, analyses with respect to the trajectories of the sensor locations are provided to remedy possible grating lobes due to the gaping positions of the transmitter devices.
transducers. The state of the art for synthetic aperture sonar (SAS) is strongly coupled to constraints, especially with respect to the trajectory of the placements and the need for good navigation data. In this article, we introduce an approach to SAS using registration of scans from single arrays, i.e., at individual poses of arbitrary trajectories, hence avoiding the need for navigation data of conventional SAS systems. The approach is introduced here for the near field using the coherent phase information of sonar scans. A Delay and Sum (D&S) beamformer (BF) is used, which directly operates on pixel/voxel form on a Cartesian grid supporting the registration. It is shown that this pixel/voxel-based registration and the coherent processing of several scans forming a synthetic aperture yields substantial improvements of the image resolution. The experimental evaluation is done with an advanced simulation tool generating realistic 2D sonar array data, i.e., with simulations of a linear 1D antenna reconstructing 2D images. For the image registration of the raw sonar scans, a robust implementation of a spectral method is presented. Furthermore, analyses with respect to the trajectories of the sensor locations are provided to remedy possible grating lobes due to the gaping positions of the transmitter devices.
Bülow, Heiko; Mueller, Christian Atanas; Buda, Frederike; Chavez, Arturo Gomez; Birk, Andreas
A Divide and Conquer Method for 3D Registration of Inhomogeneous, Partially Overlapping Scans with Fourier Mellin SOFT (FMS) Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), 2020.
@inproceedings{Valentin3D-FMS-DivideNConquer-ICRA20,
title = {A Divide and Conquer Method for 3D Registration of Inhomogeneous, Partially Overlapping Scans with Fourier Mellin SOFT (FMS)},
author = {Heiko Bülow and Christian Atanas Mueller and Frederike Buda and Arturo Gomez Chavez and Andreas Birk},
url = {https://www.researchgate.net/publication/344982408_A_Divide_and_Conquer_Method_for_3D_Registration_of_Inhomogeneous_Partially_Overlapping_Scans_with_Fourier_Mellin_SOFT_FMS},
year = {2020},
date = {2020-01-01},
urldate = {2020-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
abstract = {High-end laser range-finders provide accurate 3D data over long ranges. But their scans are inhomogeneous, i.e., the environment is non-uniformly sampled, as there is denser data in the near range than in the far range. Furthermore, the generation of a scan is time-consuming. Thus, it is desirable to cover an area by as few scans as possible, i.e., scanning is more time-efficient if the overlap between scans is as small as possible. However, these factors pose significant challenges for state-of-the-art registration algorithms. In this work, we present a divide-and-conquer method that uses an efficient strategy to check for possible registrations between partitions of two scans. As underlying registration method, Fourier-Mellin-SOFT (FMS) is used. FMS is quite robust against partial overlaps but its performance is significantly boosted by the presented partitioning method. As concrete use case, results from the digitization of a WWII submarine bunker as a large-scale cultural heritage site are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
High-end laser range-finders provide accurate 3D data over long ranges. But their scans are inhomogeneous, i.e., the environment is non-uniformly sampled, as there is denser data in the near range than in the far range. Furthermore, the generation of a scan is time-consuming. Thus, it is desirable to cover an area by as few scans as possible, i.e., scanning is more time-efficient if the overlap between scans is as small as possible. However, these factors pose significant challenges for state-of-the-art registration algorithms. In this work, we present a divide-and-conquer method that uses an efficient strategy to check for possible registrations between partitions of two scans. As underlying registration method, Fourier-Mellin-SOFT (FMS) is used. FMS is quite robust against partial overlaps but its performance is significantly boosted by the presented partitioning method. As concrete use case, results from the digitization of a WWII submarine bunker as a large-scale cultural heritage site are presented.
Mueller, Christian A.; Birk, Andreas
Visual Object Categorization Based on Hierarchical Shape Motifs Learned From Noisy Point Cloud Decompositions Journal Article
In: Journal of Intelligent and Robotic Systems, vol. 97, no. 2, pp. 313-338, 2020, ISSN: 1573-0409.
@article{ShapeMotifs-JIRS20,
title = {Visual Object Categorization Based on Hierarchical Shape Motifs Learned From Noisy Point Cloud Decompositions},
author = {Christian A. Mueller and Andreas Birk},
url = {https://doi.org/10.1007/s10846-019-01016-y},
doi = {10.1007/s10846-019-01016-y},
issn = {1573-0409},
year = {2020},
date = {2020-01-01},
journal = {Journal of Intelligent and Robotic Systems},
volume = {97},
number = {2},
pages = {313-338},
abstract = {Object shape is a key cue that contributes to the semantic understanding of objects. In this work we focus on the categorization of real-world object point clouds to particular shape types. Therein surface description and representation of object shape structure have significant influence on shape categorization accuracy, when dealing with real-world scenes featuring noisy, partial and occluded object observations. An unsupervised hierarchical learning procedure is utilized here to symbolically describe surface characteristics on multiple semantic levels. Furthermore, a constellation model is proposed that hierarchically decomposes objects. The decompositions are described as constellations of symbols (shape motifs) in a gradual order, hence reflecting shape structure from local to global, i.e., from parts over groups of parts to entire objects. The combination of this multi-level description of surfaces and the hierarchical decomposition of shapes leads to a representation which allows to conceptualize shapes. An object discrimination has been observed in experiments with seven categories featuring instances with sensor noise, occlusions as well as inter-category and intra-category similarities. Experiments include the evaluation of the proposed description and shape decomposition approach, and comparisons to Fast Point Feature Histograms, a Vocabulary Tree and a neural network-based Deep Learning method. Furthermore, experiments are conducted with alternative datasets which analyze the generalization capability of the proposed approach.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Object shape is a key cue that contributes to the semantic understanding of objects. In this work we focus on the categorization of real-world object point clouds to particular shape types. Therein surface description and representation of object shape structure have significant influence on shape categorization accuracy, when dealing with real-world scenes featuring noisy, partial and occluded object observations. An unsupervised hierarchical learning procedure is utilized here to symbolically describe surface characteristics on multiple semantic levels. Furthermore, a constellation model is proposed that hierarchically decomposes objects. The decompositions are described as constellations of symbols (shape motifs) in a gradual order, hence reflecting shape structure from local to global, i.e., from parts over groups of parts to entire objects. The combination of this multi-level description of surfaces and the hierarchical decomposition of shapes leads to a representation which allows to conceptualize shapes. An object discrimination has been observed in experiments with seven categories featuring instances with sensor noise, occlusions as well as inter-category and intra-category similarities. Experiments include the evaluation of the proposed description and shape decomposition approach, and comparisons to Fast Point Feature Histograms, a Vocabulary Tree and a neural network-based Deep Learning method. Furthermore, experiments are conducted with alternative datasets which analyze the generalization capability of the proposed approach.
Mueller, Christian A.; Chavez, Arturo Gomez; Doernbach, Tobias; Köhntopp, Daniel; Birk, Andreas
Continuous system integration and validation for underwater perception in offshore inspection and intervention tasks Book Chapter
In: Karimi, Hamid Reza (Ed.): Fundamental Design and Automation Technologies in Offshore Robotics, Chapter 2, pp. 9-75, Elsevier, 2020, ISBN: 978-0-12-820271-5.
@inbook{DexROV-ContinuousIntegration-BookChapter20,
title = {Continuous system integration and validation for underwater perception in offshore inspection and intervention tasks},
author = {Christian A. Mueller and Arturo Gomez Chavez and Tobias Doernbach and Daniel Köhntopp and Andreas Birk},
editor = {Hamid Reza Karimi},
url = {https://www.sciencedirect.com/science/article/pii/B9780128202715000079},
doi = {https://doi.org/10.1016/B978-0-12-820271-5.00007-9},
isbn = {978-0-12-820271-5},
year = {2020},
date = {2020-01-01},
booktitle = {Fundamental Design and Automation Technologies in Offshore Robotics},
pages = {9-75},
publisher = {Elsevier},
chapter = {2},
abstract = {Continuous System Integration and Validation is an increasingly important factor for an efficient system development process. In particular, for underwater projects involving semi- to fully-autonomous robotic systems since they progressively become more complex, need to perform under more challenging environmental conditions and execute more intricate tasks. Therefore, a benchmark and validation framework become crucial to quantify the system performance, as well as to identify and analyze uncertainties, bottlenecks, limitations, etc., before field trials and missions. We present a simulation-driven framework whose concepts ease the development of benchmark and validation tests that are of particular interest in interdisciplinary projects with multiple contributors at different locations. While considering aspects related to continuous system integration and validation, the presented framework facilitates a more efficient and subsequently effective project realization. Considering the preparation efforts of such system, these pay off as they allow for fast informed development cycles and qualified assessments, minimizing the integration labour on-site at field missions which usually involve high running costs and time constraints. The effectiveness of the mentioned framework is demonstrated through the development and analysis of the EU-H2020 DexROV project.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Continuous System Integration and Validation is an increasingly important factor for an efficient system development process. In particular, for underwater projects involving semi- to fully-autonomous robotic systems since they progressively become more complex, need to perform under more challenging environmental conditions and execute more intricate tasks. Therefore, a benchmark and validation framework become crucial to quantify the system performance, as well as to identify and analyze uncertainties, bottlenecks, limitations, etc., before field trials and missions. We present a simulation-driven framework whose concepts ease the development of benchmark and validation tests that are of particular interest in interdisciplinary projects with multiple contributors at different locations. While considering aspects related to continuous system integration and validation, the presented framework facilitates a more efficient and subsequently effective project realization. Considering the preparation efforts of such system, these pay off as they allow for fast informed development cycles and qualified assessments, minimizing the integration labour on-site at field missions which usually involve high running costs and time constraints. The effectiveness of the mentioned framework is demonstrated through the development and analysis of the EU-H2020 DexROV project.
2019
Chavez, Arturo Gomez; Mueller, Christian A.; Doernbach, Tobias; Birk, Andreas
Underwater Navigation using Visual Markers in the Context of Intervention Missions Journal Article
In: International Journal of Advanced Robotic Systems (IJARS), 2019.
@article{DexROV-NavigationMarkers-IJARS19,
title = {Underwater Navigation using Visual Markers in the Context of Intervention Missions},
author = {Arturo Gomez Chavez and Christian A. Mueller and Tobias Doernbach and Andreas Birk},
url = {https://www.researchgate.net/publication/332611500_Underwater_navigation_using_visual_markers_in_the_context_of_intervention_missions},
doi = {10.1177/1729881419838967},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
journal = {International Journal of Advanced Robotic Systems (IJARS)},
abstract = {Intervention missions, that is, underwater manipulation tasks, for example, in the context of oil-&-gas production, require a high amount of precise, robust navigation. In this article, we describe the use of an advanced vision system suited for deep-sea operations, which in combination with artificial markers on target structures like oil-&-gas production-Christmas-trees significantly boosts navigation performance. The system is validated in two intensive field tests running off the shore of Marseille, France. In the experiments, a commercial remotely operated vehicle equipped with the system and a mock-up structure with an oil-&-gas production panel is used to evaluate the navigation performance.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Intervention missions, that is, underwater manipulation tasks, for example, in the context of oil-&-gas production, require a high amount of precise, robust navigation. In this article, we describe the use of an advanced vision system suited for deep-sea operations, which in combination with artificial markers on target structures like oil-&-gas production-Christmas-trees significantly boosts navigation performance. The system is validated in two intensive field tests running off the shore of Marseille, France. In the experiments, a commercial remotely operated vehicle equipped with the system and a mock-up structure with an oil-&-gas production panel is used to evaluate the navigation performance.
Chavez, Arturo Gomez; Ranieri, Andrea; Chiarella, Davide; Zereik, Enrica; Babic, Anja; Birk, Andreas
CADDY Underwater Stereo-Vision Dataset for Human-Robot Interaction (HRI) in the Context of Diver Activities Journal Article
In: Journal of Marine Science and Engineering (JMSE), spec.iss. Underwater Imaging, vol. 7, no. 1, 2019.
@article{CADDYdataset-JMSE19,
title = {CADDY Underwater Stereo-Vision Dataset for Human-Robot Interaction (HRI) in the Context of Diver Activities},
author = {Arturo Gomez Chavez and Andrea Ranieri and Davide Chiarella and Enrica Zereik and Anja Babic and Andreas Birk},
url = {https://www.mdpi.com/2077-1312/7/1/16/pdf?version=1547637387},
doi = {10.3390/jmse7010016},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
journal = {Journal of Marine Science and Engineering (JMSE), spec.iss. Underwater Imaging},
volume = {7},
number = {1},
abstract = {In this article, we present a novel underwater dataset collected from several field trials within the EU FP7 project “Cognitive autonomous diving buddy (CADDY)”, where an Autonomous Underwater Vehicle (AUV) was used to interact with divers and monitor their activities. To our knowledge, this is one of the first efforts to collect a large public dataset in underwater environments with the purpose of studying and boosting object classification, segmentation and human pose estimation tasks. The first part of the dataset contains stereo camera recordings (≈10 K) of divers performing hand gestures to communicate with an AUV in different environmental conditions. The gestures can be used to test the robustness of visual detection and classification algorithms in underwater conditions, e.g., under color attenuation and light backscatter. The second part includes stereo footage (≈12.7 K) of divers free-swimming in front of the AUV, along with synchronized measurements from Inertial Measurement Units (IMU) located throughout the diver’s suit (DiverNet), which serve as ground-truth for human pose and tracking methods. In both cases, these rectified images allow the investigation of 3D representation and reasoning pipelines from low-texture targets commonly present in underwater scenarios. This work describes the recording platform, sensor calibration procedure plus the data format and the software utilities provided to use the dataset.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this article, we present a novel underwater dataset collected from several field trials within the EU FP7 project “Cognitive autonomous diving buddy (CADDY)”, where an Autonomous Underwater Vehicle (AUV) was used to interact with divers and monitor their activities. To our knowledge, this is one of the first efforts to collect a large public dataset in underwater environments with the purpose of studying and boosting object classification, segmentation and human pose estimation tasks. The first part of the dataset contains stereo camera recordings (≈10 K) of divers performing hand gestures to communicate with an AUV in different environmental conditions. The gestures can be used to test the robustness of visual detection and classification algorithms in underwater conditions, e.g., under color attenuation and light backscatter. The second part includes stereo footage (≈12.7 K) of divers free-swimming in front of the AUV, along with synchronized measurements from Inertial Measurement Units (IMU) located throughout the diver’s suit (DiverNet), which serve as ground-truth for human pose and tracking methods. In both cases, these rectified images allow the investigation of 3D representation and reasoning pipelines from low-texture targets commonly present in underwater scenarios. This work describes the recording platform, sensor calibration procedure plus the data format and the software utilities provided to use the dataset.
Chavez, Arturo Gomez; Xu, Qingwen; Mueller, Christian Atanas; Schwertfeger, Sören; Birk, Andreas
Adaptive Navigation Scheme for Optimal Deep-Sea Localization Using Multimodal Perception Cues Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019.
@inproceedings{DexROV-ORBfeatures-Navigation-IROS19,
title = {Adaptive Navigation Scheme for Optimal Deep-Sea Localization Using Multimodal Perception Cues},
author = {Arturo Gomez Chavez and Qingwen Xu and Christian Atanas Mueller and Sören Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/338946474_Adaptive_Navigation_Scheme_for_Optimal_Deep-Sea_Localization_Using_Multimodal_Perception_Cues},
doi = {10.1109/IROS40897.2019.8967888},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
abstract = {Underwater robot interventions require a high level of safety and reliability. A major challenge to address is a robust and accurate acquisition of localization estimates, as it is a prerequisite to enable more complex tasks, e.g. floating manipulation and mapping. State-of-the-art navigation in commercial operations, such as oil& gas production (OGP), rely on costly instrumentation. These can be partially replaced or assisted by visual navigation methods, especially in deep-sea scenarios where equipment deployment has high costs and risks. Our work presents a multimodal approach that adapts state-of-the-art methods from on-land robotics, i.e., dense point cloud generation in combination with plane representation and registration, to boost underwater localization performance. A (a) two-stage navigation scheme is proposed that initially generates a coarse probabilistic map of the workspace, which is used to filter noise from computed point clouds and planes in the second stage. Furthermore, an adaptive decision-making approach is introduced that determines which perception cues to incorporate into the localization filter to optimize accuracy and computation performance. Our approach is investigated first in simulation and then validated with data from field trials in OGP monitoring and maintenance scenarios.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater robot interventions require a high level of safety and reliability. A major challenge to address is a robust and accurate acquisition of localization estimates, as it is a prerequisite to enable more complex tasks, e.g. floating manipulation and mapping. State-of-the-art navigation in commercial operations, such as oil& gas production (OGP), rely on costly instrumentation. These can be partially replaced or assisted by visual navigation methods, especially in deep-sea scenarios where equipment deployment has high costs and risks. Our work presents a multimodal approach that adapts state-of-the-art methods from on-land robotics, i.e., dense point cloud generation in combination with plane representation and registration, to boost underwater localization performance. A (a) two-stage navigation scheme is proposed that initially generates a coarse probabilistic map of the workspace, which is used to filter noise from computed point clouds and planes in the second stage. Furthermore, an adaptive decision-making approach is introduced that determines which perception cues to incorporate into the localization filter to optimize accuracy and computation performance. Our approach is investigated first in simulation and then validated with data from field trials in OGP monitoring and maintenance scenarios.
Chavez, Arturo Gomez; Xu, Qingwen; Mueller, Christian A.; Schwertfeger, Soren; Birk, Andreas
Towards Accurate Deep-Sea Localization in Structured Environments based on Perception Quality Cues Proceedings Article
In: Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS), pp. 1988-1990, International Foundation for Autonomous Agents and Multiagent Systems, 3331986, 2019.
@inproceedings{DexROV-PerceptionLocalization-Qingwen-AAMAS19,
title = {Towards Accurate Deep-Sea Localization in Structured Environments based on Perception Quality Cues},
author = {Arturo Gomez Chavez and Qingwen Xu and Christian A. Mueller and Soren Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/336953145_Towards_Accurate_Deep-Sea_Localization_in_Structured_Environments_Based_on_Perception_Quality_Cues},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
booktitle = {Proceedings of the 18th International Conference on Autonomous Agents and MultiAgent Systems (AAMAS)},
pages = {1988-1990},
publisher = {International Foundation for Autonomous Agents and Multiagent Systems},
address = {3331986},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Köhntopp, Daniel; Lehmann, Benjamin; Kraus, Dieter; Birk, Andreas
Classification and Localization of Naval Mines with Superellipse Active Contours Journal Article
In: IEEE Journal of Oceanic Engineering, vol. 44, no. 3, pp. 767-782, 2019, (online first: 2018).
@article{MineClassificationLocalization-Superellipses-JOE19,
title = {Classification and Localization of Naval Mines with Superellipse Active Contours},
author = {Daniel Köhntopp and Benjamin Lehmann and Dieter Kraus and Andreas Birk},
url = {https://www.researchgate.net/publication/325837031_Classification_and_Localization_of_Naval_Mines_With_Superellipse_Active_Contours},
doi = {10.1109/JOE.2018.2835218},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
journal = {IEEE Journal of Oceanic Engineering},
volume = {44},
number = {3},
pages = {767-782},
abstract = {In this paper, an approach for the classification and localization of geometric shapes, e.g., man-made objects or different types of geological features, in sonar images is presented. It is applied to a concrete application case, namely the detection and classification of naval mines. The approach consists of three steps. In the first step, the sonar image is segmented by a new active contours algorithm. To deal with the significant noise on sonar images, the assumption is used that the segmenting contours of highlight and shadow areas of objects are geometric shapes that can be described by superellipses. It is shown here that this superellipse constraint, which can cover a wide range from box-shaped to round objects, can directly be incorporated into an active contours method without an additional edge framework. In addition to its robustness to noise, our superellipse-driven active contours approach has the advantage that it is adaptable to the intensity distribution properties of sonar images. The second step consists of the actual classification including a pose estimation using a standard naive Bayes classifier on the superellipse parameters that are computed by the segmentation in the first step. Robustness is further boosted in a novel third step in which the classification is verified in a top-down process. Based on the results of the bottom-up processes, i.e., the segmentation in step one and the pose estimation from the superellipse parameters plus the class estimation by the classifier in step two, it is possible to simulate what the input sonar image should look like if the results are correct. If this model-based top-down simulation is similar to the original sensor image, the classification result is accepted; otherwise it is rejected. To this end, different measures are compared to compute the similarity between the real sensor image and the anticipated image generated from the classification and localization hypothesis. Finally, our approach is evaluated with a challenging real-world data set of 213 synthetic aperture sonar sidescan images from sea trials with mock-up mines.},
note = {online first: 2018},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In this paper, an approach for the classification and localization of geometric shapes, e.g., man-made objects or different types of geological features, in sonar images is presented. It is applied to a concrete application case, namely the detection and classification of naval mines. The approach consists of three steps. In the first step, the sonar image is segmented by a new active contours algorithm. To deal with the significant noise on sonar images, the assumption is used that the segmenting contours of highlight and shadow areas of objects are geometric shapes that can be described by superellipses. It is shown here that this superellipse constraint, which can cover a wide range from box-shaped to round objects, can directly be incorporated into an active contours method without an additional edge framework. In addition to its robustness to noise, our superellipse-driven active contours approach has the advantage that it is adaptable to the intensity distribution properties of sonar images. The second step consists of the actual classification including a pose estimation using a standard naive Bayes classifier on the superellipse parameters that are computed by the segmentation in the first step. Robustness is further boosted in a novel third step in which the classification is verified in a top-down process. Based on the results of the bottom-up processes, i.e., the segmentation in step one and the pose estimation from the superellipse parameters plus the class estimation by the classifier in step two, it is possible to simulate what the input sonar image should look like if the results are correct. If this model-based top-down simulation is similar to the original sensor image, the classification result is accepted; otherwise it is rejected. To this end, different measures are compared to compute the similarity between the real sensor image and the anticipated image generated from the classification and localization hypothesis. Finally, our approach is evaluated with a challenging real-world data set of 213 synthetic aperture sonar sidescan images from sea trials with mock-up mines.
Luczynski, Tomasz; Luczynski, Piotr; LukasPehle,; Wirsum, Manfred; Birk, Andreas
Model based design of a stereo vision system for intelligent deep-sea operations Journal Article
In: Measurement, vol. 144, pp. 298-310, 2019.
@article{DexROV-StereoDesign-Measurement2019,
title = {Model based design of a stereo vision system for intelligent deep-sea operations},
author = {Tomasz Luczynski and Piotr Luczynski and LukasPehle and Manfred Wirsum and Andreas Birk},
url = {https://www.researchgate.net/publication/332992241_Model_Based_Design_of_a_Stereo_Vision_System_for_Intelligent_Deep-Sea_Operations},
doi = {10.1016/j.measurement.2019.05.004},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
journal = {Measurement},
volume = {144},
pages = {298-310},
abstract = {A stereo vision system for deep-sea operations is presented. The system consists of cameras in pressure bottles, which are daisy-chained to a computer bottle. The system has substantial computation power for on-board stereo processing as well as for further computer vision methods to support autonomous intelligent functions, e.g., object recognition, navigation, mapping, inspection, and intervention. The model based design presented here includes two main aspects. First, a formalized approach to the component selection for the stereo set-up is introduced, i.e., given especially accuracy and baseline constraints as well as lens and imager options, an algorithmic analysis is provided. This approach is also of interest for the design of stereo systems in general. Second, the specific aspects of deep sea operations are addressed. This includes especially the validation and optimization of the pressure bottles for the cameras with a Finite Element Method (FEM). Experiments and results are presented, which include a validation of the stereo performance in air, robustness tests of the bottles in pressure tanks, and field trials of the complete system off the shore of Marseille on a commercial Remotely Operated Vehicle (ROV).},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A stereo vision system for deep-sea operations is presented. The system consists of cameras in pressure bottles, which are daisy-chained to a computer bottle. The system has substantial computation power for on-board stereo processing as well as for further computer vision methods to support autonomous intelligent functions, e.g., object recognition, navigation, mapping, inspection, and intervention. The model based design presented here includes two main aspects. First, a formalized approach to the component selection for the stereo set-up is introduced, i.e., given especially accuracy and baseline constraints as well as lens and imager options, an algorithmic analysis is provided. This approach is also of interest for the design of stereo systems in general. Second, the specific aspects of deep sea operations are addressed. This includes especially the validation and optimization of the pressure bottles for the cameras with a Finite Element Method (FEM). Experiments and results are presented, which include a validation of the stereo performance in air, robustness tests of the bottles in pressure tanks, and field trials of the complete system off the shore of Marseille on a commercial Remotely Operated Vehicle (ROV).
Vaskevicius, Narunas; Birk, Andreas
Revisiting Superquadric Fitting: A numerically stable formulation Journal Article
In: IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 1, pp. 220-233, 2019, ISSN: 0162-8828.
@article{SuperquadricFitting-TPAMI19,
title = {Revisiting Superquadric Fitting: A numerically stable formulation},
author = {Narunas Vaskevicius and Andreas Birk},
url = {https://www.researchgate.net/publication/321506977_Revisiting_Superquadric_Fitting_A_Numerically_Stable_Formulation},
doi = {10.1109/TPAMI.2017.2779493},
issn = {0162-8828},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
volume = {41},
number = {1},
pages = {220-233},
abstract = {Superquadric surfaces play an important role in many different research fields due to their ability to model a variety of shapes with a small number of parameters. One of their core applications is the estimation of object shape characteristics from a set of discrete samples from the surface of an object. However, the corresponding optimization problem is prone to numerical instabilities in some regions of the parameter space. To mitigate this problem, lower bound constraints to the shape parameters are applied during the optimization thus limiting the range of shapes which can be accurately represented by superquadrics. Therefore, the exact modeling of very common shapes like cuboids and cylinders is error-prone in practice. In this article we investigate this problem and provide a numerically stable formulation for the evaluation of the superquadric surface function and for its gradient. This new formulation enables numerically stable fitting of superquadrics in the previously constrained region, i.e., in its full range including cuboids and cylinders. In addition, the new formulation also leads to faster convergence speed. The theoretical contributions are substantiated by experiments on synthetic as well as real data.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Superquadric surfaces play an important role in many different research fields due to their ability to model a variety of shapes with a small number of parameters. One of their core applications is the estimation of object shape characteristics from a set of discrete samples from the surface of an object. However, the corresponding optimization problem is prone to numerical instabilities in some regions of the parameter space. To mitigate this problem, lower bound constraints to the shape parameters are applied during the optimization thus limiting the range of shapes which can be accurately represented by superquadrics. Therefore, the exact modeling of very common shapes like cuboids and cylinders is error-prone in practice. In this article we investigate this problem and provide a numerically stable formulation for the evaluation of the superquadric surface function and for its gradient. This new formulation enables numerically stable fitting of superquadrics in the previously constrained region, i.e., in its full range including cuboids and cylinders. In addition, the new formulation also leads to faster convergence speed. The theoretical contributions are substantiated by experiments on synthetic as well as real data.
Xu, Qingwen; Chavez, Arturo Gomez; Bulow, Heiko; Birk, Andreas; Schwertfeger, Sören
Improved Fourier Mellin Invariant for Robust Rotation Estimation with Omni-Cameras Proceedings Article
In: IEEE International Conference on Image Processing (ICIP), pp. 320-324, 2019.
@inproceedings{iFMI-OmniCamera-ICIP19,
title = {Improved Fourier Mellin Invariant for Robust Rotation Estimation with Omni-Cameras},
author = {Qingwen Xu and Arturo Gomez Chavez and Heiko Bulow and Andreas Birk and Sören Schwertfeger},
url = {https://www.researchgate.net/publication/328939351_Improved_Fourier_Mellin_Invariant_for_Robust_Rotation_Estimation_with_Omni-cameras},
doi = {10.1109/icip.2019.8802933},
year = {2019},
date = {2019-01-01},
urldate = {2019-01-01},
booktitle = {IEEE International Conference on Image Processing (ICIP)},
pages = {320-324},
abstract = {Spectral methods such as the improved Fourier Mellin Invariant (iFMI) transform have proved to be faster, more robust and accurate than feature based methods on image registration. However, iFMI is restricted to work only when the camera moves in 2D space and has not been applied on omni-cameras images so far. In this work, we extend the iFMI method and apply a motion model to estimate an omni-camera’s pose when it moves in 3D space. In the experiment section, we compare the extended iFMI method against ORB and AKAZE feature based approaches on three datasets, showing different types of environments: office, lawn and urban scenery (MPI-omni dataset). The results show that our method reduces the error of the camera pose estimation two to three times with respect to the feature registration techniques, while offering lower processing times.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Spectral methods such as the improved Fourier Mellin Invariant (iFMI) transform have proved to be faster, more robust and accurate than feature based methods on image registration. However, iFMI is restricted to work only when the camera moves in 2D space and has not been applied on omni-cameras images so far. In this work, we extend the iFMI method and apply a motion model to estimate an omni-camera’s pose when it moves in 3D space. In the experiment section, we compare the extended iFMI method against ORB and AKAZE feature based approaches on three datasets, showing different types of environments: office, lawn and urban scenery (MPI-omni dataset). The results show that our method reduces the error of the camera pose estimation two to three times with respect to the feature registration techniques, while offering lower processing times.
2018
Bülow, Heiko; Birk, Andreas
Scale-Free Registrations in 3D: 7 Degrees of Freedom with Fourier-Mellin-SOFT transforms Journal Article
In: International Journal of Computer Vision (IJCV), vol. 126, no. 7, pp. 731-750, 2018.
@article{Fourier-Mellin-SOFT-FMS-7DoFregistration-IJCV18,
title = {Scale-Free Registrations in 3D: 7 Degrees of Freedom with Fourier-Mellin-SOFT transforms},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/323361408_Scale-Free_Registrations_in_3D_7_Degrees_of_Freedom_with_Fourier_Mellin_SOFT_Transforms},
doi = {10.1007/s11263-018-1067-5. 2018},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
journal = {International Journal of Computer Vision (IJCV)},
volume = {126},
number = {7},
pages = {731-750},
abstract = {Fourier Mellin SOFT (FMS) as a novel method for global registration of 3D data is presented. It determines the seven degrees of freedom (7-DoF) transformation, i.e., the 6-DoF rigid motion parameters plus 1-DoF scale, between two scans, i.e., two noisy, only partially overlapping views on objects or scenes. It is based on a sequence of the 3D Fourier transform, the Mellin transform and the SO(3) Fourier transform. This combination represents a non-trivial complete 3D extension of the well known Fourier-Mellin registration for 2D images. It is accordingly based on decoupling rotation and scale from translation. First, rotation—which is the main challenge for the extension to 3D data - is tackled with a SO(3) Fourier Transform (SOFT) based on Spherical Harmonics. In a second step, scale is determined via a 3D Mellin transform. Finally, translation is calculated by Phase-Matching. Experiments are presented with simulated data sets for ground truth comparisons and with real world data including object recognition and localization in Magnetic Resonance Tomography (MRT) data, registration of 2.5D RGBD scans from a Microsoft Kinect with a scale-free 3D model generated by Multi-View Vision, and 3D mapping by registration of a sequence of consecutive scans from a low-cost actuated Laser Range Finder. The results show that the method is fast and that it can robustly handle partial overlap, interfering structures, and noise. It is also shown that the method is a very interesting option for 6-DoF registration, i.e., when scale is known.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Fourier Mellin SOFT (FMS) as a novel method for global registration of 3D data is presented. It determines the seven degrees of freedom (7-DoF) transformation, i.e., the 6-DoF rigid motion parameters plus 1-DoF scale, between two scans, i.e., two noisy, only partially overlapping views on objects or scenes. It is based on a sequence of the 3D Fourier transform, the Mellin transform and the SO(3) Fourier transform. This combination represents a non-trivial complete 3D extension of the well known Fourier-Mellin registration for 2D images. It is accordingly based on decoupling rotation and scale from translation. First, rotation—which is the main challenge for the extension to 3D data - is tackled with a SO(3) Fourier Transform (SOFT) based on Spherical Harmonics. In a second step, scale is determined via a 3D Mellin transform. Finally, translation is calculated by Phase-Matching. Experiments are presented with simulated data sets for ground truth comparisons and with real world data including object recognition and localization in Magnetic Resonance Tomography (MRT) data, registration of 2.5D RGBD scans from a Microsoft Kinect with a scale-free 3D model generated by Multi-View Vision, and 3D mapping by registration of a sequence of consecutive scans from a low-cost actuated Laser Range Finder. The results show that the method is fast and that it can robustly handle partial overlap, interfering structures, and noise. It is also shown that the method is a very interesting option for 6-DoF registration, i.e., when scale is known.
Birk, A.; Doernbach, T.; Mueller, C. A.; Luczynski, T.; Chavez, A. Gomez; Köhntopp, D.; Kupcsik, A.; Calinon, S.; Tanwani, A. K.; Antonelli, G.; di Lillo, P.; Simetti, E.; Casalino, G.; Indiveri, G.; Ostuni, L.; Turetta, A.; Caffaz, A.; Weiss, P.; Gobert, T.; Chemisky, B.; Gancet, J.; Siedel, T.; Govindaraj, S.; Martinez, X.; Letier, P.
Dexterous Underwater Manipulation from Distant Onshore Locations Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 25, no. 4, pp. 24-33, 2018.
@article{DexROV-FirstFieldTrials-RAM18,
title = {Dexterous Underwater Manipulation from Distant Onshore Locations},
author = {A. Birk and T. Doernbach and C. A. Mueller and T. Luczynski and A. Gomez Chavez and D. Köhntopp and A. Kupcsik and S. Calinon and A. K. Tanwani and G. Antonelli and P. di Lillo and E. Simetti and G. Casalino and G. Indiveri and L. Ostuni and A. Turetta and A. Caffaz and P. Weiss and T. Gobert and B. Chemisky and J. Gancet and T. Siedel and S. Govindaraj and X. Martinez and P. Letier},
url = {https://www.researchgate.net/publication/328156038_Dexterous_Underwater_Manipulation_from_Onshore_Locations_Streamlining_Efficiencies_for_Remotely_Operated_Underwater_Vehicles},
doi = {10.1109/MRA.2018.2869523},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {25},
number = {4},
pages = {24-33},
abstract = {Underwater manipulation is a challenging problem. The state-of-the-art technology is dominated by remotely operated vehicles (ROVs). ROV operations typically require an offshore crew consisting of, at minimum, an intendant (or supervisor), an operator, and a navigator. This crew must often be doubled or even tripled due to work shifts. In addition, customer representatives often wish to be physically present offshore. Furthermore, underwater intervention missions are still dominated by a significant amount of lowlevel, manual control of the manipulator(s) and of the vehicle itself. While there is a significant amount of research on autonomous underwater vehicles (AUVs) in general and fieldable solutions already exist for inspection and exploration missions, possibilities remain for adding intelligent autonomous functions for interventions.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Underwater manipulation is a challenging problem. The state-of-the-art technology is dominated by remotely operated vehicles (ROVs). ROV operations typically require an offshore crew consisting of, at minimum, an intendant (or supervisor), an operator, and a navigator. This crew must often be doubled or even tripled due to work shifts. In addition, customer representatives often wish to be physically present offshore. Furthermore, underwater intervention missions are still dominated by a significant amount of lowlevel, manual control of the manipulator(s) and of the vehicle itself. While there is a significant amount of research on autonomous underwater vehicles (AUVs) in general and fieldable solutions already exist for inspection and exploration missions, possibilities remain for adding intelligent autonomous functions for interventions.
Doernbach, Tobias; Chavez, Arturo Gomez; Mueller, Christian A.; Birk, Andreas
High-Fidelity Deep-Sea Perception Using Simulation in the Loop Proceedings Article
In: IFAC Conference on Control Applications in Marine Systems (CAMS), 2018.
@inproceedings{DexROV-PerceptionSimulation-CAMS18,
title = {High-Fidelity Deep-Sea Perception Using Simulation in the Loop},
author = {Tobias Doernbach and Arturo Gomez Chavez and Christian A. Mueller and Andreas Birk},
url = {https://www.researchgate.net/publication/328242547_High-Fidelity_Deep-Sea_Perception_Using_Simulation_in_the_Loop},
doi = {10.1016/j.ifacol.2018.09.465},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
booktitle = {IFAC Conference on Control Applications in Marine Systems (CAMS)},
abstract = {Deep-sea robotic operations require a high level of safety, efficiency and reliability. In the development stage of such systems, measures have to be taken into account to validate performance in order to assess the achievement of these requirements. In the context of continuous system integration, we proposed a simulation-in-the-loop framework focusing on the mitigation of discrepancies between simulation and real-world conditions. While in our previous work we mainly targeted a high-fidelity simulation that embeds spatial conditions from recorded real-world data, this work emphasizes environmental conditions. We propose an optimization cycle which allows to enhance the fidelity of simulated underwater camera images in a backward optimization step and to enhance real-world images with knowledge available in simulation in a forward optimization step. Experimental results show that the proposed methodology optimized both simulation and real imagery, and subsequently ensures high fidelity.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deep-sea robotic operations require a high level of safety, efficiency and reliability. In the development stage of such systems, measures have to be taken into account to validate performance in order to assess the achievement of these requirements. In the context of continuous system integration, we proposed a simulation-in-the-loop framework focusing on the mitigation of discrepancies between simulation and real-world conditions. While in our previous work we mainly targeted a high-fidelity simulation that embeds spatial conditions from recorded real-world data, this work emphasizes environmental conditions. We propose an optimization cycle which allows to enhance the fidelity of simulated underwater camera images in a backward optimization step and to enhance real-world images with knowledge available in simulation in a forward optimization step. Experimental results show that the proposed methodology optimized both simulation and real imagery, and subsequently ensures high fidelity.
Mueller, Christian Atanas; Birk, Andreas
Conceptualization of Object Compositions Using Persistent Homology Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018.
@inproceedings{ObjectCompositionsPersistentHomology-IROS18,
title = {Conceptualization of Object Compositions Using Persistent Homology},
author = {Christian Atanas Mueller and Andreas Birk},
url = {https://www.researchgate.net/publication/328232451_Conceptualization_of_Object_Compositions_Using_Persistent_Homology},
doi = {10.1109/IROS.2018.8594516},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
abstract = {A topological shape analysis is proposed and utilized to learn concepts that reflect shape commonalities. Our approach is two-fold: i) a spatial topology analysis of point cloud segment constellations within objects. Therein constellations are decomposed and described in an hierarchical manner – from single segments to segment groups until a single group reflects an entire object. ii) a topology analysis of the description space in which segment decompositions are exposed in. Inspired by Persistent Homology, hidden groups of shape commonalities are revealed from object segment decompositions. Experiments show that extracted persistent groups of commonalities can represent semantically meaningful shape concepts. We also show the generalization capability of the proposed approach considering samples of external datasets.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A topological shape analysis is proposed and utilized to learn concepts that reflect shape commonalities. Our approach is two-fold: i) a spatial topology analysis of point cloud segment constellations within objects. Therein constellations are decomposed and described in an hierarchical manner – from single segments to segment groups until a single group reflects an entire object. ii) a topology analysis of the description space in which segment decompositions are exposed in. Inspired by Persistent Homology, hidden groups of shape commonalities are revealed from object segment decompositions. Experiments show that extracted persistent groups of commonalities can represent semantically meaningful shape concepts. We also show the generalization capability of the proposed approach considering samples of external datasets.
Mueller, Christian Atanas; Doernbach, Tobias; Chavez, Arturo Gomez; Köhntopp, Daniel; Birk, Andreas
Robust Continuous System Integration for Critical Deep-Sea Robot Operations Using Knowledge-Enabled Simulation in the Loop Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, 2018.
@inproceedings{DexROV-SimulationInTheLoop-IROS18,
title = {Robust Continuous System Integration for Critical Deep-Sea Robot Operations Using Knowledge-Enabled Simulation in the Loop},
author = {Christian Atanas Mueller and Tobias Doernbach and Arturo Gomez Chavez and Daniel Köhntopp and Andreas Birk},
url = {https://www.researchgate.net/publication/323597580_Robust_Continuous_System_Integration_for_Critical_Deep-Sea_Robot_Operations_Using_Knowledge-Enabled_Simulation_in_the_Loop},
doi = {10.1109/IROS.2018.8594392},
year = {2018},
date = {2018-01-01},
urldate = {2018-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE},
abstract = {Deep-sea robot operations demand a high level of safety, efficiency and reliability. As a consequence, measures within the development stage have to be implemented to extensively evaluate and benchmark system components ranging from data acquisition, perception and localization to control. We present an approach based on high-fidelity simulation that embeds spatial and environmental conditions from recorded real-world data. This simulation in the loop (SIL) methodology allows for mitigating the discrepancy between simulation and real-world conditions, e.g. regarding sensor noise. As a result, this work provides a platform to thoroughly investigate and benchmark behaviors of system components concurrently under real and simulated conditions. The conducted evaluation shows the benefit of the proposed work in tasks related to perception and self-localization under changing spatial and environmental conditions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deep-sea robot operations demand a high level of safety, efficiency and reliability. As a consequence, measures within the development stage have to be implemented to extensively evaluate and benchmark system components ranging from data acquisition, perception and localization to control. We present an approach based on high-fidelity simulation that embeds spatial and environmental conditions from recorded real-world data. This simulation in the loop (SIL) methodology allows for mitigating the discrepancy between simulation and real-world conditions, e.g. regarding sensor noise. As a result, this work provides a platform to thoroughly investigate and benchmark behaviors of system components concurrently under real and simulated conditions. The conducted evaluation shows the benefit of the proposed work in tasks related to perception and self-localization under changing spatial and environmental conditions.
2017
Benndorf, Maik; Haenselmann, Thomas; Garsch, Maximilian; Gebbeken, Norbert; Mueller, Christian A.; Fromm, Tobias; Luczynski, Tomasz; Birk, Andreas
Robotic bridge statics assessment within strategic flood evacuation planning using low-cost sensors Proceedings Article
In: IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR), 2017.
@inproceedings{SSRR17-BridgeInspection-VibrationSmartphones,
title = {Robotic bridge statics assessment within strategic flood evacuation planning using low-cost sensors},
author = {Maik Benndorf and Thomas Haenselmann and Maximilian Garsch and Norbert Gebbeken and Christian A. Mueller and Tobias Fromm and Tomasz Luczynski and Andreas Birk},
url = {https://www.researchgate.net/publication/320373248_Robotic_Bridge_Statics_Assessment_Within_Strategic_Flood_Evacuation_Planning_Using_Low-Cost_Sensors},
doi = {10.1109/SSRR.2017.8088133},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE International Symposium on Safety, Security and Rescue Robotics (SSRR)},
abstract = {Scenario: A rescue team needs to cross a partially damaged bridge in a flooded area. It is unknown whether the construction is still able to carry a vehicle. Assessing the construction's integrity can be accomplished by the analysis of the bridge's eigenfrequencies. Rather than using proprietary expensive Vibration Measurement Systems (VMS) we propose to utilize off-the-shelf smartphones as sensors-which still require to be placed at the spot on the bridge best suited for picking up vibrations. Within this work, we use an Unmanned Ground Vehicle (UGV) featuring a robotic manipulator. It allows a non-technician operator to optimally place the device semi-automatically. We evaluate our approach in a real-life scenario. Demo video: https://youtu.be/u_3pe0nZ5tw},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Scenario: A rescue team needs to cross a partially damaged bridge in a flooded area. It is unknown whether the construction is still able to carry a vehicle. Assessing the construction's integrity can be accomplished by the analysis of the bridge's eigenfrequencies. Rather than using proprietary expensive Vibration Measurement Systems (VMS) we propose to utilize off-the-shelf smartphones as sensors-which still require to be placed at the spot on the bridge best suited for picking up vibrations. Within this work, we use an Unmanned Ground Vehicle (UGV) featuring a robotic manipulator. It allows a non-technician operator to optimally place the device semi-automatically. We evaluate our approach in a real-life scenario. Demo video: https://youtu.be/u_3pe0nZ5tw
Birk, Andreas
The game of science Book Chapter
In: Belpaeme, Tony; Beuls, Katrien; Boer, Bart De (Ed.): Luc Steels - Growing Seeds, pp. 64-67, Berger and Montag Publishers, Antwerpen, 2017, ISBN: 9789081430616.
@inbook{GameOfSciene-LucLiberAmoricum17,
title = {The game of science},
author = {Andreas Birk},
editor = {Tony Belpaeme and Katrien Beuls and Bart De Boer},
isbn = {9789081430616},
year = {2017},
date = {2017-01-01},
booktitle = {Luc Steels - Growing Seeds},
pages = {64-67},
publisher = {Berger and Montag Publishers},
address = {Antwerpen},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Chavez, Arturo Gomez; Mueller, Christian A.; Birk, Andreas; Babic, Anja; Miskovic, Nikola
Stereo-vision based diver pose estimation using LSTM recurrent neural networks for AUV navigation guidance Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{CADDY-DiverPoseEstimation-OceansAberdeen17,
title = {Stereo-vision based diver pose estimation using LSTM recurrent neural networks for AUV navigation guidance},
author = {Arturo Gomez Chavez and Christian A. Mueller and Andreas Birk and Anja Babic and Nikola Miskovic},
url = {https://www.researchgate.net/publication/318432335_Stereo-vision_based_diver_pose_estimation_using_LSTM_recurrent_neural_networks_for_AUV_navigation_guidance},
doi = {10.1109/OCEANSE.2017.8085020},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {Within the EU FP7 project " Cognitive autonomous diving buddy (CADDY) " , work has been made to assist and monitor divers through Autonomous Underwater Vehicles (AUVs) during their long underwater expeditions. To achieve this goal, one milestone is to give the AUV the capability to track the diver's whereabouts at all times. Inertial sensors are mounted on the diver's body to transmit acoustically his relative position and orientation to the AUV using an Ultra-Short Baseline (USBL) system. However, the acoustic system is prone to give erroneous data depending on the surrounding geography, modems alignment and sensors calibration, plus its low transmission rate does not allow real time performance. To overcome these drawbacks and complement the acoustic setup , this paper presents a general framework to detect and estimate the diver's body pose based on generated point clouds from a stereo camera. Our method is based on a Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) that captures the temporal relations between global point cloud descriptors as they change with the diver movements. Since the analysis is made on time sequences rather than on one-shot visual information, the framework is robust against the distortions and poor quality typical in underwater while still providing real time performance. We focus on the description of the image-processing and LSTM-RNN pipeline, as well on its validation with a dataset created during several field trial experiments.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Within the EU FP7 project " Cognitive autonomous diving buddy (CADDY) " , work has been made to assist and monitor divers through Autonomous Underwater Vehicles (AUVs) during their long underwater expeditions. To achieve this goal, one milestone is to give the AUV the capability to track the diver's whereabouts at all times. Inertial sensors are mounted on the diver's body to transmit acoustically his relative position and orientation to the AUV using an Ultra-Short Baseline (USBL) system. However, the acoustic system is prone to give erroneous data depending on the surrounding geography, modems alignment and sensors calibration, plus its low transmission rate does not allow real time performance. To overcome these drawbacks and complement the acoustic setup , this paper presents a general framework to detect and estimate the diver's body pose based on generated point clouds from a stereo camera. Our method is based on a Long Short-Term Memory Recurrent Neural Network (LSTM-RNN) that captures the temporal relations between global point cloud descriptors as they change with the diver movements. Since the analysis is made on time sequences rather than on one-shot visual information, the framework is robust against the distortions and poor quality typical in underwater while still providing real time performance. We focus on the description of the image-processing and LSTM-RNN pipeline, as well on its validation with a dataset created during several field trial experiments.
Fromm, Tobias; Mueller, Christian A.; Pfingsthorn, Max; Birk, Andreas; Lillo, Paolo Di
Efficient Continuous System Integration and Validation for Deep-Sea Robotics Applications Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{DexROV-simulator-OceansAberdeen17,
title = {Efficient Continuous System Integration and Validation for Deep-Sea Robotics Applications},
author = {Tobias Fromm and Christian A. Mueller and Max Pfingsthorn and Andreas Birk and Paolo Di Lillo},
url = {https://www.researchgate.net/publication/318340259_Efficient_Continuous_System_Integration_and_Validation_for_Deep-Sea_Robotics_Applications},
doi = {10.1109/OCEANSE.2017.8084663},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {Deep-sea operations of remotely-operated vehicles (ROV) need robust testing and deployment strategies beyond the traditional pre-deployment validation on real hardware. Seamless integration of simulated components into the validation pipeline allows for rapid development of components and validation under controlled conditions. We describe the benefits arising from such a continuous integration and validation approach as well as an example setup in the EU project DexROV.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Deep-sea operations of remotely-operated vehicles (ROV) need robust testing and deployment strategies beyond the traditional pre-deployment validation on real hardware. Seamless integration of simulated components into the validation pipeline allows for rapid development of components and validation under controlled conditions. We describe the benefits arising from such a continuous integration and validation approach as well as an example setup in the EU project DexROV.
Koehntopp, Daniel; Lehmann, Benjamin; Kraus, Dieter; Birk, Andreas
Seafloor Classification for Mine Counter Measure Operations using Synthetic Aperture Sonar Image Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{SeafloorClassification-ATLAS-Demining-OceansAberdeen17,
title = {Seafloor Classification for Mine Counter Measure Operations using Synthetic Aperture Sonar Image},
author = {Daniel Koehntopp and Benjamin Lehmann and Dieter Kraus and Andreas Birk},
url = {https://www.researchgate.net/publication/318722168_Seafloor_Classification_for_Mine_Countermeasures_Operations_using_Synthetic_Aperture_Sonar_Images},
doi = {10.1109/OCEANSE.2017.8084752},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {The performance of an automatic target recognition (ATR) system in the context of naval mine detection is severely affected by the underwater environment. Especially in regions with the presence of sand ripples or mine-sized stones the number of false alarms can become unacceptable high, if the detection algorithm does not account for the type of seafloor. Therefore, a robust way of discriminating between the three most important types of seafloor in the context of mine countermeasures (MCM), flat bottom, rocky bottom and sand ripples, is needed. In this paper five handcrafted features for seafloor classification from the literature are analysed and compared to the performance of a convolutional neural network (CNN) that learns the relevant features from the training data. The evaluation data was collected with an autonomous underwater vehicle (AUV) equipped with an synthetic aperture sonar. Experiments showed that the CNN approach outperformed a support vector machines clas-sifier with the handcrafted features, albeit slightly, with a classification rate of 98.7% over 9420 examples.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The performance of an automatic target recognition (ATR) system in the context of naval mine detection is severely affected by the underwater environment. Especially in regions with the presence of sand ripples or mine-sized stones the number of false alarms can become unacceptable high, if the detection algorithm does not account for the type of seafloor. Therefore, a robust way of discriminating between the three most important types of seafloor in the context of mine countermeasures (MCM), flat bottom, rocky bottom and sand ripples, is needed. In this paper five handcrafted features for seafloor classification from the literature are analysed and compared to the performance of a convolutional neural network (CNN) that learns the relevant features from the training data. The evaluation data was collected with an autonomous underwater vehicle (AUV) equipped with an synthetic aperture sonar. Experiments showed that the CNN approach outperformed a support vector machines clas-sifier with the handcrafted features, albeit slightly, with a classification rate of 98.7% over 9420 examples.
Luczynski, Tomasz; Birk, Andreas
Underwater Image Haze Removal with an Underwater-ready Dark Channel Prior Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{HazeRemoval-DarkChanelPrior-OceansAnchorage17,
title = {Underwater Image Haze Removal with an Underwater-ready Dark Channel Prior},
author = {Tomasz Luczynski and Andreas Birk},
url = {https://www.researchgate.net/publication/320145382_Underwater_Image_Haze_Removal_with_an_Underwater-ready_Dark_Channel_Prior},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {Underwater images suffer from extremely unfavourable conditions. Light is heavily attenuated and scattered. Attenuation creates change in hue, scattering causes so called veiling light. General state of the art methods for enhancing image quality are either unreliable or cannot be easily used in underwater operations. On the other hand there is a well known method for haze removal in air, called Dark Channel Prior. Even though there are known adaptations of this method to underwater applications, they do not always work correctly. In our work a modification to the Dark Channel Prior is proposed that allows an easy application to underwater images. It is also shown that our method outperforms competing solutions and that it is a very promising solution worth further development.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater images suffer from extremely unfavourable conditions. Light is heavily attenuated and scattered. Attenuation creates change in hue, scattering causes so called veiling light. General state of the art methods for enhancing image quality are either unreliable or cannot be easily used in underwater operations. On the other hand there is a well known method for haze removal in air, called Dark Channel Prior. Even though there are known adaptations of this method to underwater applications, they do not always work correctly. In our work a modification to the Dark Channel Prior is proposed that allows an easy application to underwater images. It is also shown that our method outperforms competing solutions and that it is a very promising solution worth further development.
Luczynski, Tomasz; Fromm, Tobias; Govindaraj, Shashank; Mueller, Christian A.; Birk, Andreas
3D Grid Map Transmission for Underwater Mapping and Visualization under Bandwidth Constraints Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{DexROV-OctomapTransmission-OceansAnchorage17,
title = {3D Grid Map Transmission for Underwater Mapping and Visualization under Bandwidth Constraints},
author = {Tomasz Luczynski and Tobias Fromm and Shashank Govindaraj and Christian A. Mueller and Andreas Birk},
url = {https://www.researchgate.net/publication/320145293_3D_Grid_Map_Transmission_for_Underwater_Mapping_and_Visualization_under_Bandwidth_Constraints},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {ROV operations are expensive and there is an urge to improve on their efficiency and speed. The EU project " Effective Dexterous ROV Operations in Presence of Communications Latencies (DexROV) " proposes a solution to those needs by developing a set of hardware and software tools to support the teleoperation of ROVs over a satellite link, i.e., from remote locations including offshore sites. A core challenge is the significant amount of data that must be transmitted to the distant Mission Control Center to allow this. To ease operation in general and especially in this application case involving latencies in the control, a 3D model of the environment that is generated during the mission by sensors on the ROV is to be transmitted. This paper focuses on using a 3D grid map for this purpose and especially its efficient transmission under significant bandwidth constraints.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
ROV operations are expensive and there is an urge to improve on their efficiency and speed. The EU project " Effective Dexterous ROV Operations in Presence of Communications Latencies (DexROV) " proposes a solution to those needs by developing a set of hardware and software tools to support the teleoperation of ROVs over a satellite link, i.e., from remote locations including offshore sites. A core challenge is the significant amount of data that must be transmitted to the distant Mission Control Center to allow this. To ease operation in general and especially in this application case involving latencies in the control, a 3D model of the environment that is generated during the mission by sensors on the ROV is to be transmitted. This paper focuses on using a 3D grid map for this purpose and especially its efficient transmission under significant bandwidth constraints.
Luczynski, Tomasz; Pfingsthorn, Max; Birk, Andreas
Image Rectification with the Pinax Camera Model in Underwater Stereo Systems with Verged Cameras Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{PinaxVergedStereo-OceansAnchorage17,
title = {Image Rectification with the Pinax Camera Model in Underwater Stereo Systems with Verged Cameras},
author = {Tomasz Luczynski and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/320145466_Image_Rectification_with_the_Pinax_Camera_Model_in_Underwater_Stereo_Systems_with_Verged_Cameras},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {Vision methods are very commonly used in underwater robotics. However, the cameras must be enclosed in a sealed housing for which flat-pane windows are the most economical and hence most popular solution. But they cause non-trivial refraction-based distortions at the air-glass-water interface. Recently, a new model for underwater cameras calibration called Pinax was introduced, which combines aspects of an axial and a pinhole camera model to derive an as accurate as possible and at the same time computational feasible solution. Here it is shown, how the Pinax model can be applied to stereo cameras when two verged cameras are behind a single flat glass panel. Real world experiments under different water conditions show that our method outperforms standard methods using pinhole models in combination with in-water calibration.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Vision methods are very commonly used in underwater robotics. However, the cameras must be enclosed in a sealed housing for which flat-pane windows are the most economical and hence most popular solution. But they cause non-trivial refraction-based distortions at the air-glass-water interface. Recently, a new model for underwater cameras calibration called Pinax was introduced, which combines aspects of an axial and a pinhole camera model to derive an as accurate as possible and at the same time computational feasible solution. Here it is shown, how the Pinax model can be applied to stereo cameras when two verged cameras are behind a single flat glass panel. Real world experiments under different water conditions show that our method outperforms standard methods using pinhole models in combination with in-water calibration.
Luczynski, Tomasz; Pfingsthorn, Max; Birk, Andreas
The Pinax-Model for Accurate and Efficient Refraction Correction of Underwater Cameras in Flat-Pane Housings Journal Article
In: Ocean Engineering, vol. 133, pp. 9-22, 2017.
@article{Pinax-OceanEngineering17,
title = {The Pinax-Model for Accurate and Efficient Refraction Correction of Underwater Cameras in Flat-Pane Housings},
author = {Tomasz Luczynski and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/313417323_The_Pinax-model_for_accurate_and_efficient_refraction_correction_of_underwater_cameras_in_flat-pane_housings},
doi = {10.1016/j.oceaneng.2017.01.029},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
journal = {Ocean Engineering},
volume = {133},
pages = {9-22},
abstract = {The calibration and refraction correction process for underwater cameras with flat-pane interfaces is presented that is very easy and convenient to use in real world applications while yielding very accurate results. The correction is derived from an analysis of the axial camera model for underwater cameras, which is among others computationally hard to tackle. It is shown how realistic constraints on the distance of the camera to the window can be exploited, which leads to an approach dubbed Pinax Model as it combines aspects of a virtual pinhole model with the projection function from the axial camera model. It allows the pre-computation of a lookup-table for very fast refraction correction of the flat-pane with high accuracy. The model takes the refraction indices of water into account, especially with respect to salinity, and it is therefore sufficient to calibrate the underwater camera only once in air. It is demonstrated by real world experiments with several underwater cameras in different salt and sweet water conditions that the proposed process outperforms standard methods. Among others, it is shown how the presented method leads to accurate results with single in-air calibration and even with just estimated salinity values.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The calibration and refraction correction process for underwater cameras with flat-pane interfaces is presented that is very easy and convenient to use in real world applications while yielding very accurate results. The correction is derived from an analysis of the axial camera model for underwater cameras, which is among others computationally hard to tackle. It is shown how realistic constraints on the distance of the camera to the window can be exploited, which leads to an approach dubbed Pinax Model as it combines aspects of a virtual pinhole model with the projection function from the axial camera model. It allows the pre-computation of a lookup-table for very fast refraction correction of the flat-pane with high accuracy. The model takes the refraction indices of water into account, especially with respect to salinity, and it is therefore sufficient to calibrate the underwater camera only once in air. It is demonstrated by real world experiments with several underwater cameras in different salt and sweet water conditions that the proposed process outperforms standard methods. Among others, it is shown how the presented method leads to accurate results with single in-air calibration and even with just estimated salinity values.
Miskovic, N.; Pascoal, A.; Bibuli, M.; Caccia, M.; Neasham, J. A.; Birk, A.; Egi, M.; Grammer, K.; Marroni, A.; Vasilijevic, A.; Nad, D.; Vukic, Z.
CADDY project, year 3: The final validation trials Proceedings Article
In: OCEANS, pp. 1-5, IEEE, 2017.
@inproceedings{CADDY-Year3-Oceans17,
title = {CADDY project, year 3: The final validation trials},
author = {N. Miskovic and A. Pascoal and M. Bibuli and M. Caccia and J. A. Neasham and A. Birk and M. Egi and K. Grammer and A. Marroni and A. Vasilijevic and D. Nad and Z. Vukic},
doi = {10.1109/OCEANSE.2017.8084715},
year = {2017},
date = {2017-01-01},
booktitle = {OCEANS},
pages = {1-5},
publisher = {IEEE},
abstract = {This paper described the basic concept behind the second and final validation trials of the Cognitive Autonomous Diving Buddy (CADDY) system. The CADDY system is composed of three components: autonomous surface vehicle (ASV), an autonomous underwater vehicle (AUV) named BUDDY, and a diver. CADDY aims to establish an innovative setup between a diver and companion autonomous robots (underwater and surface) that exhibit cognitive behaviour by adapting to diver's physical state, and actions, thereby improving overall diving experience by assisting the diver. Although certain problems were encountered during trials, the collected data and experience were valuable to identify potential improvements. Most importantly, divers felt safe and comfortable working and interacting with a robotic BUDDY thus achieving the primary goal.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper described the basic concept behind the second and final validation trials of the Cognitive Autonomous Diving Buddy (CADDY) system. The CADDY system is composed of three components: autonomous surface vehicle (ASV), an autonomous underwater vehicle (AUV) named BUDDY, and a diver. CADDY aims to establish an innovative setup between a diver and companion autonomous robots (underwater and surface) that exhibit cognitive behaviour by adapting to diver's physical state, and actions, thereby improving overall diving experience by assisting the diver. Although certain problems were encountered during trials, the collected data and experience were valuable to identify potential improvements. Most importantly, divers felt safe and comfortable working and interacting with a robotic BUDDY thus achieving the primary goal.
Mueller, Christian A.; Fromm, Tobias; Buelow, Heiko; Birk, Andreas; Garsch, Maximilian; Gebbeken, Norbert
Robotic Bridge Inspection Within Strategic Flood Evacuation Planning Proceedings Article
In: IEEE Oceans, IEEE press, 2017.
@inproceedings{FloodEvac-BridgeInspectionFirstResults-OceansAberdeen17,
title = {Robotic Bridge Inspection Within Strategic Flood Evacuation Planning},
author = {Christian A. Mueller and Tobias Fromm and Heiko Buelow and Andreas Birk and Maximilian Garsch and Norbert Gebbeken},
url = {https://www.researchgate.net/publication/318340374_Robotic_Bridge_Inspection_Within_Strategic_Flood_Evacuation_Planning},
doi = {10.1109/OCEANSE.2017.8084668},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE press},
abstract = {Bridges are fragile transport infrastructure elements which require special attention in evacuation planning within flood disaster scenarios. In this paper we present work on a bridge inspection procedure with a marine vehicle in which a bridge statics analysis expert assesses the condition of the bridge from a safe operating station. Given bridge construction knowledge beforehand, our work shows how the assessment can benefit from an unmanned marine vehicle which provides information about the current structural condition of the bridge. Due to quality degradation of standard RGB camera readings caused by varying visibility conditions under water, visual information is captured from an imaging sonar. In contrary to conventional approaches, the assessment of bridges gains quality by considering live stream of sonar images which reveal bridge structure information even under rough water conditions. Additionally, an image registration method is applied which creates a map of the live stream and provides a valuable information source for further bridge assessment. We describe the deployment of the system in a real test field.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Bridges are fragile transport infrastructure elements which require special attention in evacuation planning within flood disaster scenarios. In this paper we present work on a bridge inspection procedure with a marine vehicle in which a bridge statics analysis expert assesses the condition of the bridge from a safe operating station. Given bridge construction knowledge beforehand, our work shows how the assessment can benefit from an unmanned marine vehicle which provides information about the current structural condition of the bridge. Due to quality degradation of standard RGB camera readings caused by varying visibility conditions under water, visual information is captured from an imaging sonar. In contrary to conventional approaches, the assessment of bridges gains quality by considering live stream of sonar images which reveal bridge structure information even under rough water conditions. Additionally, an image registration method is applied which creates a map of the live stream and provides a valuable information source for further bridge assessment. We describe the deployment of the system in a real test field.
Vaskevicius, Narunas; Birk, Andreas
Revisiting Superquadric Fitting: A numerically stable formulation Journal Article
In: IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), online first, pp. 1-14, 2017, ISSN: 0162-8828.
@article{SuperquadricFitting-TPAMI17-efirst,
title = {Revisiting Superquadric Fitting: A numerically stable formulation},
author = {Narunas Vaskevicius and Andreas Birk},
doi = {10.1109/tpami.2017.2779493},
issn = {0162-8828},
year = {2017},
date = {2017-01-01},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), online first},
pages = {1-14},
abstract = {Superquadric surfaces play an important role in many different research fields due to their ability to model a variety of shapes with a small number of parameters. One of their core applications is the estimation of object shape characteristics from a set of discrete samples from the surface of an object. However, the corresponding optimization problem is prone to numerical instabilities in some regions of the parameter space. To mitigate this problem, lower bound constraints to the shape parameters are applied during the optimization thus limiting the range of shapes which can be accurately represented by superquadrics. Therefore, the exact modeling of very common shapes like cuboids and cylinders is error-prone in practice. In this article we investigate this problem and provide a numerically stable formulation for the evaluation of the superquadric surface function and for its gradient. This new formulation enables numerically stable fitting of superquadrics in the previously constrained region, i.e., in its full range including cuboids and cylinders. In addition, the new formulation also leads to faster convergence speed. The theoretical contributions are substantiated by experiments on synthetic as well as real data.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Superquadric surfaces play an important role in many different research fields due to their ability to model a variety of shapes with a small number of parameters. One of their core applications is the estimation of object shape characteristics from a set of discrete samples from the surface of an object. However, the corresponding optimization problem is prone to numerical instabilities in some regions of the parameter space. To mitigate this problem, lower bound constraints to the shape parameters are applied during the optimization thus limiting the range of shapes which can be accurately represented by superquadrics. Therefore, the exact modeling of very common shapes like cuboids and cylinders is error-prone in practice. In this article we investigate this problem and provide a numerically stable formulation for the evaluation of the superquadric surface function and for its gradient. This new formulation enables numerically stable fitting of superquadrics in the previously constrained region, i.e., in its full range including cuboids and cylinders. In addition, the new formulation also leads to faster convergence speed. The theoretical contributions are substantiated by experiments on synthetic as well as real data.
Vaskevicius, Narunas; Pathak, Kaustubh; Birk, Andreas
In: Journal of Intelligent and Robotic Systems, vol. 88, no. 1, pp. 57-71, 2017.
@article{RoblogIndustrial-SuperquadricSacks-JINT17,
title = {Recognition and Localization of Sacks for Autonomous Container Unloading by Fitting Superquadrics in Noisy, Partial Views from a Low-cost RGBD Sensor},
author = {Narunas Vaskevicius and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/317216808_Recognition_and_Localization_of_Sacks_for_Autonomous_Container_Unloading_by_Fitting_Superquadrics_in_Noisy_Partial_Views_from_a_Low-cost_RGBD_Sensor},
doi = {10.1007/s10846-017-0540-7},
year = {2017},
date = {2017-01-01},
urldate = {2017-01-01},
journal = {Journal of Intelligent and Robotic Systems},
volume = {88},
number = {1},
pages = {57-71},
abstract = {A significant amount of cargo worldwide is transported in sacks and bags e.g. wheat, rice, coffee and cacao beans, etc. Despite being very strenuous and the health risks involved, the handling of sacks in logistics is predominantly done through manual labor. Hence, the automation of tasks such as cargo unloading from shipping containers is of high importance. However, it faces many challenges due to the unstructured nature of packaging. One of the prerequisites for creating autonomous systems for handling bags or sacks is a robust perception component. In this work, we present a perception pipeline to recognize and localize sacks with a low-cost sensor in unstructured settings with partial views. The backbone of our perception strategy is based on two main contributions presented in this work. First, we introduce a fast convexity test between neighboring patches, which is a part of a two-level segmentation leading to a robust detection of object candidates. Second, we formulate a numerically stable form of superquadric fitting, which allows for an extension of the feasible region of the corresponding optimization problem. Both of the contributions are of interest for applications using superquadrics for representing curved object parts and hence extend beyond the specific scenario of sack/bag recognition and localization presented here. The perception modules introduced in this work are embedded into a newly designed robotic platform capable of manipulating 70 kg sacks - a standard weight when transporting coffee and cocao beans. Moreover, the robot is fully integrated in a coffee storage warehouse. Therefore we substantiate our approach with experiments in a real-world scenario of autonomous unloading of coffee cargo delivered in a shipping container.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A significant amount of cargo worldwide is transported in sacks and bags e.g. wheat, rice, coffee and cacao beans, etc. Despite being very strenuous and the health risks involved, the handling of sacks in logistics is predominantly done through manual labor. Hence, the automation of tasks such as cargo unloading from shipping containers is of high importance. However, it faces many challenges due to the unstructured nature of packaging. One of the prerequisites for creating autonomous systems for handling bags or sacks is a robust perception component. In this work, we present a perception pipeline to recognize and localize sacks with a low-cost sensor in unstructured settings with partial views. The backbone of our perception strategy is based on two main contributions presented in this work. First, we introduce a fast convexity test between neighboring patches, which is a part of a two-level segmentation leading to a robust detection of object candidates. Second, we formulate a numerically stable form of superquadric fitting, which allows for an extension of the feasible region of the corresponding optimization problem. Both of the contributions are of interest for applications using superquadrics for representing curved object parts and hence extend beyond the specific scenario of sack/bag recognition and localization presented here. The perception modules introduced in this work are embedded into a newly designed robotic platform capable of manipulating 70 kg sacks - a standard weight when transporting coffee and cocao beans. Moreover, the robot is fully integrated in a coffee storage warehouse. Therefore we substantiate our approach with experiments in a real-world scenario of autonomous unloading of coffee cargo delivered in a shipping container.
2016
Birk, Andreas; Pfingsthorn, Max
Simultaneous Localization and Mapping (SLAM) Book Chapter
In: Encyclopedia of Electrical and Electronics Engineering, Wiley, 2016.
@inbook{SLAMoverview-EncycEEE16,
title = {Simultaneous Localization and Mapping (SLAM)},
author = {Andreas Birk and Max Pfingsthorn},
url = {https://www.researchgate.net/publication/310599150_Simultaneous_Localization_and_Mapping_SLAM},
doi = {https://doi.org/10.1002/047134608X.W8322},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {Encyclopedia of Electrical and Electronics Engineering},
publisher = {Wiley},
abstract = {This article gives an overview of simultaneous localization and mapping (SLAM), that is, probabilistic methods to generate a 2D or 3D map of unknown areas under imperfect localization. The article provides a survey of the theoretical basis of SLAM as well as the core background information about the underlying techniques for implementing actual SLAM systems.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
This article gives an overview of simultaneous localization and mapping (SLAM), that is, probabilistic methods to generate a 2D or 3D map of unknown areas under imperfect localization. The article provides a survey of the theoretical basis of SLAM as well as the core background information about the underlying techniques for implementing actual SLAM systems.
Chavez, Arturo Gomez; Fontes, Jorge; Afonso, Pedro; Pfingsthorn, Max; Birk, Andreas
Automated Species Counting using a Hierarchical Classification Approach with Haar Cascades and Multi-Descriptor Random Forests Proceedings Article
In: IEEE Oceans, 2016.
@inproceedings{UnderwaterAutomatedSpeciesCounting-Oceans16,
title = {Automated Species Counting using a Hierarchical Classification Approach with Haar Cascades and Multi-Descriptor Random Forests},
author = {Arturo Gomez Chavez and Jorge Fontes and Pedro Afonso and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/297441399_Automated_Species_Counting_using_a_Hierarchical_Classification_Approach_with_Haar_Cascades_and_Multi-Descriptor_Random_Forests},
doi = {10.1109/OCEANSAP.2016.7485544},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {IEEE Oceans},
abstract = {We present a classification framework to detect underwater terrain deformations commonly caused by marine species in real time during AUV surveys. The method uses two classifiers in a hierarchical manner to overcome the fact that there are no substantial datasets in the literature about underwater terrain and the creation of one would involve a great consumption of human and technological resources. The first classifier based on Haar cascades, scans the image from an AUV stereo camera and detects all possible candidates of interest. Then, a Multi-Descriptor Random Forest is used to classify these candidates according to which species is more likely to have caused the associated terrain deformation. This classifier is able to integrate multiple descriptors to ensure the robustness of the system despite the vast amount of image distortions that happen underwater. The performance of the method is shown during a multiple AUV survey mission, where the Camera Vehicles follow the underwater ground up close. A comparison of the framework’s output with the one from marine biologists exemplifies the accuracy of the system and its usefulness while supporting activities such as marine population monitoring to preserve nursery areas.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a classification framework to detect underwater terrain deformations commonly caused by marine species in real time during AUV surveys. The method uses two classifiers in a hierarchical manner to overcome the fact that there are no substantial datasets in the literature about underwater terrain and the creation of one would involve a great consumption of human and technological resources. The first classifier based on Haar cascades, scans the image from an AUV stereo camera and detects all possible candidates of interest. Then, a Multi-Descriptor Random Forest is used to classify these candidates according to which species is more likely to have caused the associated terrain deformation. This classifier is able to integrate multiple descriptors to ensure the robustness of the system despite the vast amount of image distortions that happen underwater. The performance of the method is shown during a multiple AUV survey mission, where the Camera Vehicles follow the underwater ground up close. A comparison of the framework’s output with the one from marine biologists exemplifies the accuracy of the system and its usefulness while supporting activities such as marine population monitoring to preserve nursery areas.
Chavez, Arturo Gomez; Pfingsthorn, Max; Rathnam, Ravi; Birk, Andreas
Visual Speed Adaptation for Improved Sensor Coverage in a Multi-Vehicle Survey Mission Proceedings Article
In: IEEE Oceans, 2016.
@inproceedings{UnderwaterSpeedAdaptation-ViewPlanning-Oceans16,
title = {Visual Speed Adaptation for Improved Sensor Coverage in a Multi-Vehicle Survey Mission},
author = {Arturo Gomez Chavez and Max Pfingsthorn and Ravi Rathnam and Andreas Birk},
url = {https://www.researchgate.net/publication/297439730_Visual_Speed_Adaptation_for_Improved_Sensor_Coverage_in_a_Multi-Vehicle_Survey_Mission},
doi = {10.1109/OCEANSAP.2016.7485710},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {IEEE Oceans},
abstract = {Autonomous underwater vehicles (AUVs) often perform high-resolution survey missions. Such missions are often planned on low resolution bathymetry maps using offline coverage planning methods, e.g., using a standard lawnmower trajectory that is adapted to the coarse-resolution representation of the terrain. We present in this paper an approach to adapt the exploration online during the mission, namely by adapting the vehicle speed as a crucial parameter with which the planned survey path is tracked. The main idea is to determine online during the mission whether the currently surveyed environment part is interesting or not and to accordingly change the speed, i.e., move faster over less interesting terrain. Concretely, this paper proposes two alternative methods to compute terrain complexity metrics that can be used for adjusting the speed of a single vehicle or a formation of vehicles online during the mission. The effectiveness of the methods is shown using visual stereo survey data. A description of field trials where the methods were used to control the speed of a multi-vehicle formation during a complex survey mission further exemplifies the usefulness of the methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Autonomous underwater vehicles (AUVs) often perform high-resolution survey missions. Such missions are often planned on low resolution bathymetry maps using offline coverage planning methods, e.g., using a standard lawnmower trajectory that is adapted to the coarse-resolution representation of the terrain. We present in this paper an approach to adapt the exploration online during the mission, namely by adapting the vehicle speed as a crucial parameter with which the planned survey path is tracked. The main idea is to determine online during the mission whether the currently surveyed environment part is interesting or not and to accordingly change the speed, i.e., move faster over less interesting terrain. Concretely, this paper proposes two alternative methods to compute terrain complexity metrics that can be used for adjusting the speed of a single vehicle or a formation of vehicles online during the mission. The effectiveness of the methods is shown using visual stereo survey data. A description of field trials where the methods were used to control the speed of a multi-vehicle formation during a complex survey mission further exemplifies the usefulness of the methods.
Fromm, Tobias; Birk, Andreas
Physics-Based Damage-Aware Manipulation Strategy Planning Using Scene Dynamics Anticipation Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2016.
@inproceedings{AnticipationManipulation-IROS16,
title = {Physics-Based Damage-Aware Manipulation Strategy Planning Using Scene Dynamics Anticipation},
author = {Tobias Fromm and Andreas Birk},
url = {https://www.researchgate.net/publication/311320636_Physics-Based_Damage-Aware_Manipulation_Strategy_Planning_Using_Scene_Dynamics_Anticipation},
doi = {10.1109/IROS.2016.7759159},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {We present a damage-aware planning approach which determines the best sequence to manipulate a number of objects in a scene. This works on task-planning level, abstracts from motion planning and anticipates the dynamics of the scene using a physics simulation. Instead of avoiding interaction with the environment, we take unintended motion of other objects into account and plan manipulation sequences which minimize the potential damage. Our method can also be used as a validation measure to judge planned motions for their feasibility in terms of damage avoidance. We evaluate our approach on one industrial scenario (autonomous container unloading) and one retail scenario (shelf replenishment).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a damage-aware planning approach which determines the best sequence to manipulate a number of objects in a scene. This works on task-planning level, abstracts from motion planning and anticipates the dynamics of the scene using a physics simulation. Instead of avoiding interaction with the environment, we take unintended motion of other objects into account and plan manipulation sequences which minimize the potential damage. Our method can also be used as a validation measure to judge planned motions for their feasibility in terms of damage avoidance. We evaluate our approach on one industrial scenario (autonomous container unloading) and one retail scenario (shelf replenishment).
Gancet, Jeremi; Weiss, Peter; Antonelli, Gianluca; Pfingsthorn, Max Folkert; Calinon, Sylvain; Turetta, Alessio; Walen, Cees; Urbina, Diego; Govindaraj, Shashank; Müller, Christian Atanas; Martinez, Xavier; Fromm, Tobias; Chemisky, Bertrand; Indiveri, Giovanni; Casalino, Giuseppe; Lillo, Paolo Augusto Di; Simetti, Enrico; Palma, Daniela De; Birk, Andreas; Tanwani, Ajay; Havoutis, Ioannis; Caffaz, Andrea; Guilpain, Lisa; Letier, Pierre
Dexterous Undersea Interventions with Far Distance Onshore Supervision: the DexROV Project Proceedings Article
In: 10th IFAC Conference on Control Applications in Marine Systems (CAMS), International Federation of Automatic Control, 2016.
@inproceedings{DexROV-CAMS16,
title = {Dexterous Undersea Interventions with Far Distance Onshore Supervision: the DexROV Project},
author = {Jeremi Gancet and Peter Weiss and Gianluca Antonelli and Max Folkert Pfingsthorn and Sylvain Calinon and Alessio Turetta and Cees Walen and Diego Urbina and Shashank Govindaraj and Christian Atanas Müller and Xavier Martinez and Tobias Fromm and Bertrand Chemisky and Giovanni Indiveri and Giuseppe Casalino and Paolo Augusto Di Lillo and Enrico Simetti and Daniela De Palma and Andreas Birk and Ajay Tanwani and Ioannis Havoutis and Andrea Caffaz and Lisa Guilpain and Pierre Letier},
url = {https://www.researchgate.net/publication/309623933_Dexterous_Undersea_Interventions_with_Far_Distance_Onshore_Supervision_the_DexROV_Project},
doi = {10.1016/j.ifacol.2016.10.439},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {10th IFAC Conference on Control Applications in Marine Systems (CAMS)},
publisher = {International Federation of Automatic Control},
abstract = {The operation of a ROV requires significant off-shore dedicated manpower to handle and operate the robotic platform. In order to reduce the burden of operations, DexROV proposes to work out more cost effective and time efficient ROV operations, where manned support is in a large extent delocalized onshore (i.e. from a ROV control center), possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme makes provision for advanced dexterous manipulation capabilities, exploiting human expertise from a remote location when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial. After one year, the project specified the system architecture of the system and carried out preliminary technological trade-offs for the subsystems.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The operation of a ROV requires significant off-shore dedicated manpower to handle and operate the robotic platform. In order to reduce the burden of operations, DexROV proposes to work out more cost effective and time efficient ROV operations, where manned support is in a large extent delocalized onshore (i.e. from a ROV control center), possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme makes provision for advanced dexterous manipulation capabilities, exploiting human expertise from a remote location when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial. After one year, the project specified the system architecture of the system and carried out preliminary technological trade-offs for the subsystems.
Miskovic, Nikola; Bibuli, Marco; Birk, Andreas; Caccia, Massimo; Egi, Murat; Grammer, Karl; Marroni, Alessandro; Neasham, Jeff; Pascoal, Antonio; Vukic, Antonio Vasilijevic Zoran
CADDY - Cognitive Autonomous Diving Buddy: Two Years of Underwater Human-Robot Interaction Journal Article
In: Marine Technology Society (MTS) Journal, vol. 50, no. 4, pp. 1-13, 2016.
@article{CADDY-year2-MTS16,
title = {CADDY - Cognitive Autonomous Diving Buddy: Two Years of Underwater Human-Robot Interaction},
author = {Nikola Miskovic and Marco Bibuli and Andreas Birk and Massimo Caccia and Murat Egi and Karl Grammer and Alessandro Marroni and Jeff Neasham and Antonio Pascoal and Antonio Vasilijevic Zoran Vukic},
url = {https://www.researchgate.net/publication/307588298_CADDY-Cognitive_Autonomous_Diving_Buddy_Two_Years_of_Underwater_Human-Robot_Interaction},
doi = {10.4031/MTSJ.50.4.11},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
journal = {Marine Technology Society (MTS) Journal},
volume = {50},
number = {4},
pages = {1-13},
abstract = {Divers operate in harsh and poorlymonitored environments, inwhich the slightest unexpected disturbance, technical malfunction, or lack of attention can have catastrophic consequences. Motivated by these considerations, the “CADDY—Cognitive Autonomous Diving Buddy” FP7 project sets forth the main goal of developing a cooperative autonomous underwater robotic system to monitor and assist human divers, thus affording them increased levels of safety during the execution of challenging scientific and commercial missions. This article presents the main results obtained in the first 2 years of the project along the following main research topics: Seeing the Diver, where the focus is placed on the 3D reconstruction of a diver’s model (pose estimation and recognition of hand gestures) through remote and local sensing technologies, thus enabling behavior interpretation; Understanding the Diver, with the objective of interpreting the model and physiological measurements of the diver in order to determine the state of the diver; and Diver-Robot Cooperation and Control, aimed at investigating the interaction of the diver with underwater vehicles endowed with rich sensory motor skills, focusing on cooperative control and optimal formation with the diver as an integral part of the overall vehiclediver formation. © 2016, Marine Technology Society Journal. All rights reserved.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Divers operate in harsh and poorlymonitored environments, inwhich the slightest unexpected disturbance, technical malfunction, or lack of attention can have catastrophic consequences. Motivated by these considerations, the “CADDY—Cognitive Autonomous Diving Buddy” FP7 project sets forth the main goal of developing a cooperative autonomous underwater robotic system to monitor and assist human divers, thus affording them increased levels of safety during the execution of challenging scientific and commercial missions. This article presents the main results obtained in the first 2 years of the project along the following main research topics: Seeing the Diver, where the focus is placed on the 3D reconstruction of a diver’s model (pose estimation and recognition of hand gestures) through remote and local sensing technologies, thus enabling behavior interpretation; Understanding the Diver, with the objective of interpreting the model and physiological measurements of the diver in order to determine the state of the diver; and Diver-Robot Cooperation and Control, aimed at investigating the interaction of the diver with underwater vehicles endowed with rich sensory motor skills, focusing on cooperative control and optimal formation with the diver as an integral part of the overall vehiclediver formation. © 2016, Marine Technology Society Journal. All rights reserved.
Miskovic, Nikola; Pascoal, Antonio; Bibuli, Marco; Caccia, Massimo; Neasham, Jeffrey A.; Birk, Andreas; Egi, Murat; Grammer, Karl; Marroni, Alessandro; Vasilijevic, Antonio; Vukic, Zoran
CADDY Project, Year 2: The First Validation Trials Proceedings Article
In: 10th IFAC Conference on Control Applications in Marine Systems (CAMS), International Federation of Automatic Control, 2016.
@inproceedings{CADDY-CAMS16,
title = {CADDY Project, Year 2: The First Validation Trials},
author = {Nikola Miskovic and Antonio Pascoal and Marco Bibuli and Massimo Caccia and Jeffrey A. Neasham and Andreas Birk and Murat Egi and Karl Grammer and Alessandro Marroni and Antonio Vasilijevic and Zoran Vukic},
url = {https://www.researchgate.net/publication/309624024_CADDY_Project_Year_2_The_First_Validation_Trials},
doi = {10.1016/j.ifacol.2016.10.440},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {10th IFAC Conference on Control Applications in Marine Systems (CAMS)},
publisher = {International Federation of Automatic Control},
abstract = {"CADDY - Cognitive Autonomous Diving Buddy" is an FP7 project that that is devoted to developing a cognitive underwater robotic system that will help divers during their activities in this hazardous environment. The envisioned resulting system will play a threefold role similar to those that a human buddy diver should have: buddy "observer", buddy "slave", and buddy "guide". During the second year of the project, validation trials were organized in Croatia with the purpose of testing all developed algorithms that will enable the three roles of the CADDY system. The trials were structured in five experiments. This paper is devoted to providing a concise overview of the conducted experiments and major results.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
"CADDY - Cognitive Autonomous Diving Buddy" is an FP7 project that that is devoted to developing a cognitive underwater robotic system that will help divers during their activities in this hazardous environment. The envisioned resulting system will play a threefold role similar to those that a human buddy diver should have: buddy "observer", buddy "slave", and buddy "guide". During the second year of the project, validation trials were organized in Croatia with the purpose of testing all developed algorithms that will enable the three roles of the CADDY system. The trials were structured in five experiments. This paper is devoted to providing a concise overview of the conducted experiments and major results.
Mueller, Christian Atanas; Birk, Andreas
Hierarchical Graph-Based Discovery of Non-Primitive-Shaped Objects in Unstructured Environments Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2016.
@inproceedings{HierarchicalGraphBasedObjectDiscovery-ICRA16,
title = {Hierarchical Graph-Based Discovery of Non-Primitive-Shaped Objects in Unstructured Environments},
author = {Christian Atanas Mueller and Andreas Birk},
url = {https://www.researchgate.net/publication/303436579_Hierarchical_Graph-Based_Discovery_of_Non-Primitive-Shaped_Objects_in_Unstructured_Environments},
doi = {10.1109/ICRA.2016.7487375},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {We present a hierarchical graph-based approach for unknown object discovery in RGB-D point clouds captured with a Kinect-like sensor from unstructured scenes. A two-step approach is proposed which first extracts meaningful regions from an input scene through over-segmentation. Secondly, a procedure is introduced to detect compositions of such regions that can represent primitive-shaped object candidates like boxes or cylinders. Complex-shaped objects are interpreted as a composition of primitive-shaped objects, for instance, a teddy bear can consist of two convex-shaped arms, legs, a convex-shaped head and torso. An ensemble of classifiers is trained to learn patterns from the appearances of such neighboring primitive shapes that constitute complex-shaped objects. Therein the appearance is described by a set of features from the texture and geometry domain. For the experiments, a dataset was prepared which is publicly available, containing a set of scenes which consists of 296 human-annotated object instances in total. Experiments show that the proposed hierarchical approach is capable to extract meaningful regions: an under-segmentation rate of 2.6% has been achieved. Furthermore, objects are segmented with a segmentation rate of 92.9% which reflects the capability of our approach to detect potential object candidates within unstructured scenes.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present a hierarchical graph-based approach for unknown object discovery in RGB-D point clouds captured with a Kinect-like sensor from unstructured scenes. A two-step approach is proposed which first extracts meaningful regions from an input scene through over-segmentation. Secondly, a procedure is introduced to detect compositions of such regions that can represent primitive-shaped object candidates like boxes or cylinders. Complex-shaped objects are interpreted as a composition of primitive-shaped objects, for instance, a teddy bear can consist of two convex-shaped arms, legs, a convex-shaped head and torso. An ensemble of classifiers is trained to learn patterns from the appearances of such neighboring primitive shapes that constitute complex-shaped objects. Therein the appearance is described by a set of features from the texture and geometry domain. For the experiments, a dataset was prepared which is publicly available, containing a set of scenes which consists of 296 human-annotated object instances in total. Experiments show that the proposed hierarchical approach is capable to extract meaningful regions: an under-segmentation rate of 2.6% has been achieved. Furthermore, objects are segmented with a segmentation rate of 92.9% which reflects the capability of our approach to detect potential object candidates within unstructured scenes.
Pfingsthorn, Max; Birk, Andreas
Generalized Graph SLAM: Solving Local and Global Ambiguities through Multimodal and Hyperedge Constraints Journal Article
In: International Journal of Robotics Research (IJRR), vol. 35, no. 16, pp. 601-630, 2016.
@article{GeneralizedGraphSLAM-IJRR16,
title = {Generalized Graph SLAM: Solving Local and Global Ambiguities through Multimodal and Hyperedge Constraints},
author = {Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/281316573_Generalized_graph_SLAM_Solving_local_and_global_ambiguities_through_multimodal_and_hyperedge_constraints},
doi = {10.1177/0278364915585395},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
journal = {International Journal of Robotics Research (IJRR)},
volume = {35},
number = {16},
pages = {601-630},
abstract = {Research in Graph-based Simultaneous Localization and Mapping has experienced a recent trend towards robust methods. These methods take the combinatorial aspect of data association into account by allowing decisions of the graph topology to be made during optimization. The Generalized Graph Simultaneous Localization and Mapping framework presented in this work can represent ambiguous data on both local and global scales, i.e. it can handle multiple mutually exclusive choices in registration results and potentially erroneous loop closures. This is achieved by augmenting previous work on multimodal distributions with an extended graph structure using hyperedges to encode ambiguous loop closures. The novel representation combines both hyperedges and multimodal Mixture of Gaussian constraints to represent all sources of ambiguity in Simultaneous Localization and Mapping. Furthermore, a discrete optimization stage is introduced between the Simultaneous Localization and Mapping frontend and backend to handle these ambiguities in a unified way utilizing the novel representation of Generalized Graph Simultaneous Localization and Mapping, providing a general approach to handle all forms of outliers. The novel Generalized Prefilter method optimizes among all local and global choices and generates a traditional unimodal unambiguous pose graph for subsequent continuous optimization in the backend. Systematic experiments on synthetic datasets show that the novel representation of the Generalized Graph Simultaneous Localization and Mapping framework with the Generalized Prefilter method, is significantly more robust and faster than other robust state-of-the-art methods. In addition, two experiments with real data are presented to corroborate the results observed with synthetic data. Different general strategies to construct problems from real data, utilizing the full representational power of the Generalized Graph Simultaneous Localization and Mapping framework are also illustrated in these experiments.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Research in Graph-based Simultaneous Localization and Mapping has experienced a recent trend towards robust methods. These methods take the combinatorial aspect of data association into account by allowing decisions of the graph topology to be made during optimization. The Generalized Graph Simultaneous Localization and Mapping framework presented in this work can represent ambiguous data on both local and global scales, i.e. it can handle multiple mutually exclusive choices in registration results and potentially erroneous loop closures. This is achieved by augmenting previous work on multimodal distributions with an extended graph structure using hyperedges to encode ambiguous loop closures. The novel representation combines both hyperedges and multimodal Mixture of Gaussian constraints to represent all sources of ambiguity in Simultaneous Localization and Mapping. Furthermore, a discrete optimization stage is introduced between the Simultaneous Localization and Mapping frontend and backend to handle these ambiguities in a unified way utilizing the novel representation of Generalized Graph Simultaneous Localization and Mapping, providing a general approach to handle all forms of outliers. The novel Generalized Prefilter method optimizes among all local and global choices and generates a traditional unimodal unambiguous pose graph for subsequent continuous optimization in the backend. Systematic experiments on synthetic datasets show that the novel representation of the Generalized Graph Simultaneous Localization and Mapping framework with the Generalized Prefilter method, is significantly more robust and faster than other robust state-of-the-art methods. In addition, two experiments with real data are presented to corroborate the results observed with synthetic data. Different general strategies to construct problems from real data, utilizing the full representational power of the Generalized Graph Simultaneous Localization and Mapping framework are also illustrated in these experiments.
Pfingsthorn, Max; Rathnam, Ravi; Luczynski, Tomasz; Birk, Andreas
Full 3D Navigation Correction using Low Frequency Visual Tracking with a Stereo Camera Proceedings Article
In: IEEE Oceans, 2016.
@inproceedings{UnderwaterVisualOdometry-Oceans16,
title = {Full 3D Navigation Correction using Low Frequency Visual Tracking with a Stereo Camera},
author = {Max Pfingsthorn and Ravi Rathnam and Tomasz Luczynski and Andreas Birk},
url = {https://www.researchgate.net/publication/297437694_Full_3D_Navigation_Correction_using_Low_Frequency_Visual_Tracking_with_a_Stereo_Camera},
doi = {10.1109/OCEANSAP.2016.7485520},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {IEEE Oceans},
abstract = {Robust and accurate navigation of underwater vehicles is an essential cornerstone for mission success. Integrative techniques, e.g., using inertial navigation or doppler velocity logs (DVL), can show significant drift. Even absolute input, such as GNSS input or magnetometers, can either be biased by surrounding structures or very noisy. For good navigation performance in critical maneuvers close to underwater structures like sea mounts, cliffs or man-made constructions where DVLs fail, navigation based on visual input can be very helpful. This paper presents a technique to correct a traditional navigation solution with visual tracking feedback from a stereo camera with small computational overhead. Experiments with real data from an ATLAS SeaCat AUV shows very good navigation performance on two trajectories over a variety of 3D structures.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Robust and accurate navigation of underwater vehicles is an essential cornerstone for mission success. Integrative techniques, e.g., using inertial navigation or doppler velocity logs (DVL), can show significant drift. Even absolute input, such as GNSS input or magnetometers, can either be biased by surrounding structures or very noisy. For good navigation performance in critical maneuvers close to underwater structures like sea mounts, cliffs or man-made constructions where DVLs fail, navigation based on visual input can be very helpful. This paper presents a technique to correct a traditional navigation solution with visual tracking feedback from a stereo camera with small computational overhead. Experiments with real data from an ATLAS SeaCat AUV shows very good navigation performance on two trajectories over a variety of 3D structures.
Stoyanov, Todor; Vaskevicius, Narunas; Mueller, Christian Atanas; Fromm, Tobias; Krug, Robert; Tincani, Vinicio; Mojtahedzadeh, Rasoul; Kunaschk, Stefan; Ernits, Rafael Mortensen; Canelhas, Daniel Ricao; Bonilla, Manuel; Schwertfeger, Sören; Bonini, Marco; Halfar, Harry; Pathak, Kaustubh; Rohde, Moritz; Fantoni, Gualtiero; Bicchi, Antonio; Birk, Andreas; Lilienthal, Achim J.; Echelmeyer, Wolfgang
No More Heavy Lifting: Robotic Solutions to the Container Unloading Problem Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 23, no. 4, pp. 94-106, 2016.
@article{RoblogOverview-RAM16,
title = {No More Heavy Lifting: Robotic Solutions to the Container Unloading Problem},
author = {Todor Stoyanov and Narunas Vaskevicius and Christian Atanas Mueller and Tobias Fromm and Robert Krug and Vinicio Tincani and Rasoul Mojtahedzadeh and Stefan Kunaschk and Rafael Mortensen Ernits and Daniel Ricao Canelhas and Manuel Bonilla and Sören Schwertfeger and Marco Bonini and Harry Halfar and Kaustubh Pathak and Moritz Rohde and Gualtiero Fantoni and Antonio Bicchi and Andreas Birk and Achim J. Lilienthal and Wolfgang Echelmeyer},
url = {https://www.researchgate.net/publication/307087772_No_More_Heavy_Lifting_Robotic_Solutions_to_the_Container_Unloading_Problem},
doi = {10.1109/MRA.2016.2535098},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {23},
number = {4},
pages = {94-106},
abstract = {This article discusses the scientifically and industrially important problem of automating the process of unloading goods from standard shipping containers. We outline some of the challenges barring further adoption of robotic solutions to this problem: ranging from handling a vast variety of shapes, sizes, weights, appearance and packing arrangement of the goods, through hard demands on unloading speed and reliability, to ensuring fragile goods are not damaged. We propose a modular and reconfigurable software framework in an attempt at efficiently addressing some of these challenges. We outline the general framework design, as well as the basic functionality of the core modules developed and present two instantiations of the software system on two different fully integrated demonstrators. While one is coping with an industrial scenario, namely the automated unloading of coffee sacks, with an already econom ically interesting performance, the other scenario is used to demonstrate the capabilities of our scientific and technological developments in the context of medium- to long-term prospects of automation in logistics. We performed evaluations which allow us to summarize several important lessons learned and to identify future directions of research on autonomous robots for handling of goods in logistics applications.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This article discusses the scientifically and industrially important problem of automating the process of unloading goods from standard shipping containers. We outline some of the challenges barring further adoption of robotic solutions to this problem: ranging from handling a vast variety of shapes, sizes, weights, appearance and packing arrangement of the goods, through hard demands on unloading speed and reliability, to ensuring fragile goods are not damaged. We propose a modular and reconfigurable software framework in an attempt at efficiently addressing some of these challenges. We outline the general framework design, as well as the basic functionality of the core modules developed and present two instantiations of the software system on two different fully integrated demonstrators. While one is coping with an industrial scenario, namely the automated unloading of coffee sacks, with an already econom ically interesting performance, the other scenario is used to demonstrate the capabilities of our scientific and technological developments in the context of medium- to long-term prospects of automation in logistics. We performed evaluations which allow us to summarize several important lessons learned and to identify future directions of research on autonomous robots for handling of goods in logistics applications.
Vaskevicius, Narunas; Pathak, Kaustubh; Birk, Andreas
Anticipation and Attention for Robust Object Recognition with RGBD-Data in an Industrial Application Scenario Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2016.
@inproceedings{RoblogAnticipationAttention-IROS16,
title = {Anticipation and Attention for Robust Object Recognition with RGBD-Data in an Industrial Application Scenario},
author = {Narunas Vaskevicius and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/311757495_Anticipation_and_attention_for_robust_object_recognition_with_RGBD-data_in_an_industrial_application_scenario},
doi = {10.1109/IROS.2016.7759445},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {An extension based on attention and anticipation of a robot vision pipeline for object recognition in RGBD images from low-cost sensors like MS Kinect or ASUS Xtion is presented. This work originated in research on an industrial application scenario, namely shipping-container unloading, but it is applicable to advanced manipulation tasks in unstructured environments in general where the perception must be very robust while being as fast as possible. For these scenarios, we build on our previous work that proved to be competitive in cluttered scenes in table-top scenarios and which forms the backbone of our RGBD object recognition. It is further enhanced by two main contributions. First, a simple but very effective form of anticipation as top-down expectations of the evolution of the scene due to the actions of the robot is used to speed up the processing. Second, attention is used as a mechanism for further speed-up by focusing processing only on certain regions of interest of the scene based also on an anticipation mechanism. The method is analyzed in experiments using real-world data from an industrial demonstration set-up.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An extension based on attention and anticipation of a robot vision pipeline for object recognition in RGBD images from low-cost sensors like MS Kinect or ASUS Xtion is presented. This work originated in research on an industrial application scenario, namely shipping-container unloading, but it is applicable to advanced manipulation tasks in unstructured environments in general where the perception must be very robust while being as fast as possible. For these scenarios, we build on our previous work that proved to be competitive in cluttered scenes in table-top scenarios and which forms the backbone of our RGBD object recognition. It is further enhanced by two main contributions. First, a simple but very effective form of anticipation as top-down expectations of the evolution of the scene due to the actions of the robot is used to speed up the processing. Second, attention is used as a mechanism for further speed-up by focusing processing only on certain regions of interest of the scene based also on an anticipation mechanism. The method is analyzed in experiments using real-world data from an industrial demonstration set-up.
Winkler, Jan; Balint-Benczedi, Ferenc; Vaskevicius, Narunas; Wiedemeyer, Thiemo; Mueller, Christian A.; Fromm, Tobias; Beetz, Michael; Birk, Andreas
Knowledge-Enabled Robotic Agents for Shelf Replenishment in Cluttered Retail Environments Proceedings Article
In: International Confenference on Autonomous Agents and Multiagent Systems (AAMAS), Association for Computing Machinery (ACM), 2016.
@inproceedings{Perception-KnowledgeEnabled-Retail-AAMAS16,
title = {Knowledge-Enabled Robotic Agents for Shelf Replenishment in Cluttered Retail Environments},
author = {Jan Winkler and Ferenc Balint-Benczedi and Narunas Vaskevicius and Thiemo Wiedemeyer and Christian A. Mueller and Tobias Fromm and Michael Beetz and Andreas Birk},
url = {https://www.researchgate.net/publication/302583644_Knowledge-Enabled_Robotic_Agents_for_Shelf_Replenishment_in_Cluttered_Retail_Environments},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
booktitle = {International Confenference on Autonomous Agents and Multiagent Systems (AAMAS)},
publisher = {Association for Computing Machinery (ACM)},
abstract = {Autonomous robots in unstructured and dynamically changing retail environments have to master complex perception, knowledge processing, and manipulation tasks. To enable them to act competently, we propose a framework based on three core components: (o) a knowledge-enabled perception system, capable of combining diverse information sources to cope with occlusions and stacked objects with a variety of textures and shapes, (o) knowledge processing methods produce strategies for tidying up supermarket racks, and (o) the necessary manipulation skills in confined spaces to arrange objects in semi-accessible rack shelves. We demonstrate our framework in an simulated environment as well as on a real shopping rack using a PR2 robot. Typical supermarket products are detected and rearranged in the retail rack, tidying up what was found to be misplaced items.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Autonomous robots in unstructured and dynamically changing retail environments have to master complex perception, knowledge processing, and manipulation tasks. To enable them to act competently, we propose a framework based on three core components: (o) a knowledge-enabled perception system, capable of combining diverse information sources to cope with occlusions and stacked objects with a variety of textures and shapes, (o) knowledge processing methods produce strategies for tidying up supermarket racks, and (o) the necessary manipulation skills in confined spaces to arrange objects in semi-accessible rack shelves. We demonstrate our framework in an simulated environment as well as on a real shopping rack using a PR2 robot. Typical supermarket products are detected and rearranged in the retail rack, tidying up what was found to be misplaced items.
Kalwa, Joerg; Tietjen, Daniel; Carreiro-Silva, Marina; Fontes, Jorge; Brignone, Lorenzo; Gracias, Nuno; Ridao, Pere; Pfingsthorn, Max; Birk, Andreas; Glotzbach, Thomas; Eckstein, Sebastian; Caccia, Massimo; Alves, João; Furfaro, Thomas; Ribeiro, Jorge; Pascoal, Antonio
The European Project MORPH: Distributed UUV Systems for Multimodal, 3D Underwater Surveys Journal Article
In: Marine Technology Society Journal, vol. 50, no. 4, pp. 26 - 51, 2016.
@article{MORPH-MTSjournal16,
title = {The European Project MORPH: Distributed UUV Systems for Multimodal, 3D Underwater Surveys},
author = {Joerg Kalwa and Daniel Tietjen and Marina Carreiro-Silva and Jorge Fontes and Lorenzo Brignone and Nuno Gracias and Pere Ridao and Max Pfingsthorn and Andreas Birk and Thomas Glotzbach and Sebastian Eckstein and Massimo Caccia and João Alves and Thomas Furfaro and Jorge Ribeiro and Antonio Pascoal},
url = {https://www.researchgate.net/publication/307588607_The_european_project_MORPH_Distributed_UUV_systems_for_multimodal_3D_underwater_surveys},
doi = {10.4031/MTSJ.50.4.10},
year = {2016},
date = {2016-01-01},
urldate = {2016-01-01},
journal = {Marine Technology Society Journal},
volume = {50},
number = {4},
pages = {26 - 51},
abstract = {The MORPH project (FP 7, 2012‐2016) is aimed at developing efficient methods and tools to map the underwater environment in situations that are not easily addressed by current technology. Namely, the missions that are of interest are those that involve underwater surveying and marine habitat mapping of rugged terrain and structures with full 3D complexity, including vertical cliffs. Potential applications include the study of cold water coral reef communities, ecosystems from underwater canyons, pipeline and harbor monitoring, or the inspection of wind turbine foundations. The project introduced and advanced a novel concept of an underwater robotic system composed of a number of mobile robot modules (nodes), carrying complementary sensors for perception of the environment. Instead of being physically coupled, the modules are connected via communication links that allow a flow of essential information among them. Without rigid links, the so-called MORPH Supra-Vehicle can reconfigure itself and adapt according to the environment and mission goals, responding, for example, to the shape of the terrain, including vertical walls. The flexibility allows for more optimal positioning of each sensor, increased number of simultaneous viewpoints, and generally high-resolution data collection.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The MORPH project (FP 7, 2012‐2016) is aimed at developing efficient methods and tools to map the underwater environment in situations that are not easily addressed by current technology. Namely, the missions that are of interest are those that involve underwater surveying and marine habitat mapping of rugged terrain and structures with full 3D complexity, including vertical cliffs. Potential applications include the study of cold water coral reef communities, ecosystems from underwater canyons, pipeline and harbor monitoring, or the inspection of wind turbine foundations. The project introduced and advanced a novel concept of an underwater robotic system composed of a number of mobile robot modules (nodes), carrying complementary sensors for perception of the environment. Instead of being physically coupled, the modules are connected via communication links that allow a flow of essential information among them. Without rigid links, the so-called MORPH Supra-Vehicle can reconfigure itself and adapt according to the environment and mission goals, responding, for example, to the shape of the terrain, including vertical walls. The flexibility allows for more optimal positioning of each sensor, increased number of simultaneous viewpoints, and generally high-resolution data collection.
2015
Chavez, Arturo Gomez; Birk, Andreas
Underwater Gesture Recognition based on Multi-Descriptor Random Forests (MD-NCMF) Miscellaneous
2015.
@misc{UnderwaterGestureClassification-CADDY15,
title = {Underwater Gesture Recognition based on Multi-Descriptor Random Forests (MD-NCMF)},
author = {Arturo Gomez Chavez and Andreas Birk},
url = {https://robotics.constructor.university/projects/caddy/},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
publisher = {Progress Report: EU-FP7 Cognitive autonomous diving buddy (CADDY), grant 611373},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Chavez, Arturo Gomez; Pfingsthorn, Max; Birk, Andreas; Rendulic, Ivor; Miskovic, Nikola
Visual Diver Detection using Multi-Descriptor Nearest-Class-Mean Random Forests in the Context of Underwater Human Robot Interaction (HRI) Proceedings Article
In: IEEE Oceans, 2015.
@inproceedings{DiverDetection-MultiDescriptorNCMR-Oceans15,
title = {Visual Diver Detection using Multi-Descriptor Nearest-Class-Mean Random Forests in the Context of Underwater Human Robot Interaction (HRI)},
author = {Arturo Gomez Chavez and Max Pfingsthorn and Andreas Birk and Ivor Rendulic and Nikola Miskovic},
url = {https://www.researchgate.net/publication/308840265_Visual_diver_detection_using_multi-descriptor_nearest-class-mean_random_forests_in_the_context_of_underwater_Human_Robot_Interaction_HRI},
doi = {10.1109/OCEANS-Genova.2015.7271556},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IEEE Oceans},
abstract = {This paper introduces a visual method for diver detection in the context of Human Robot Interaction (HRI). The detection is treated as a classification problem, where a discriminative model is trained by computing image features of the target (diver) and underwater scenery. This type of scenery poses great challenges due to its high variability, as it often presents high illumination changes, scarce features and image distortions. For this reason, it is desirable to represent this type of images with multiple type of complementary features. The system scalability is, however, lowered as the number of features types increase as the amount of data to represent queries and indexes also increases. To remedy this, we modified the Nearest Class Mean Forests (NCMF) method, a variant of Random Forests, to integrate as many features types as desired without concerning about scalability and performance decay. The system outperforms the common generative tracking methods which fail to encompass different type of distortions into one model and ignore background information. And in contrast to tracking methods using acoustic sensors which output a single value (distance to the diver), our approach outputs a region encompassing the diver's body; information that can be further exploited to enhance underwater HRI. Not to mention that camera setups offer higher flexibility in size and energy consumption constraints than acoustic sensors. All of the system's aforementioned capabilities are tested with real-life data obtained from field experiments.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper introduces a visual method for diver detection in the context of Human Robot Interaction (HRI). The detection is treated as a classification problem, where a discriminative model is trained by computing image features of the target (diver) and underwater scenery. This type of scenery poses great challenges due to its high variability, as it often presents high illumination changes, scarce features and image distortions. For this reason, it is desirable to represent this type of images with multiple type of complementary features. The system scalability is, however, lowered as the number of features types increase as the amount of data to represent queries and indexes also increases. To remedy this, we modified the Nearest Class Mean Forests (NCMF) method, a variant of Random Forests, to integrate as many features types as desired without concerning about scalability and performance decay. The system outperforms the common generative tracking methods which fail to encompass different type of distortions into one model and ignore background information. And in contrast to tracking methods using acoustic sensors which output a single value (distance to the diver), our approach outputs a region encompassing the diver's body; information that can be further exploited to enhance underwater HRI. Not to mention that camera setups offer higher flexibility in size and energy consumption constraints than acoustic sensors. All of the system's aforementioned capabilities are tested with real-life data obtained from field experiments.
Enchev, Ivaylo; Pfingsthorn, Max; Luczynski, Tomasz; Sokolovski, Igor; Birk, Andreas; Tietjen, Daniel
Underwater Place Recognition in Noisy Stereo Data using Fab-Map with a Multimodal Vocabulary from 2D Texture and 3D Surface Descriptors Proceedings Article
In: IEEE Oceans, 2015.
@inproceedings{UnderwaterPlaceRecognition-Multimodal-Oceans15,
title = {Underwater Place Recognition in Noisy Stereo Data using Fab-Map with a Multimodal Vocabulary from 2D Texture and 3D Surface Descriptors},
author = {Ivaylo Enchev and Max Pfingsthorn and Tomasz Luczynski and Igor Sokolovski and Andreas Birk and Daniel Tietjen},
url = {https://www.researchgate.net/publication/283071036_Underwater_place_recognition_in_noisy_stereo_data_using_FAB-MAP_with_a_multimodal_vocabulary_from_2D_texture_and_3D_surface_descriptors},
doi = {10.1109/OCEANS-Genova.2015.7271561},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IEEE Oceans},
abstract = {One fundamental problem in Simultaneous Localization and Mapping is place recognition, used to create loop closures, necessary to improve map quality. Ideally, the sensor data alone should be used to detect loop closures, referred to as appearance-based methods. Such methods usually employ feature descriptors, especially popular in underwater research are 2D visual features, and the bag-of-words approach. Here we demonstrate the combined use of 2D visual and 3D surface features, which is specifically interesting in the context of 3D underwater mapping with stereo data. Different sensor and feature types excel at different situations, and thus would benefit from a joint treatment in a bag-of-words based place recognition method. Specifically, this paper presents a method to employ multiple dictionaries computed from different feature types, extracted from different raw data in a single FAB-MAP-based place recognition method. Experiments on an underwater dataset show an increase of performance of the method when using a combination of feature types instead of single feature types.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
One fundamental problem in Simultaneous Localization and Mapping is place recognition, used to create loop closures, necessary to improve map quality. Ideally, the sensor data alone should be used to detect loop closures, referred to as appearance-based methods. Such methods usually employ feature descriptors, especially popular in underwater research are 2D visual features, and the bag-of-words approach. Here we demonstrate the combined use of 2D visual and 3D surface features, which is specifically interesting in the context of 3D underwater mapping with stereo data. Different sensor and feature types excel at different situations, and thus would benefit from a joint treatment in a bag-of-words based place recognition method. Specifically, this paper presents a method to employ multiple dictionaries computed from different feature types, extracted from different raw data in a single FAB-MAP-based place recognition method. Experiments on an underwater dataset show an increase of performance of the method when using a combination of feature types instead of single feature types.
Gancet, J.; Urbina, D.; Letier, P.; Ilzkovitz, M.; Birk, A.; Pfingsthorn, M.; Weiss, P.; Gauch, F.; Chemisky, B.; Calinon, S.; Antonelli, G.; Casalino, G.; Indiveri, G.; Turetta, A.; Walen, C.; Guilpain, L.
DexROV: Enabling Effective Dexterous ROV Operations in Presence of Communication Latency Proceedings Article
In: IEEE Oceans, 2015.
@inproceedings{DexROV-Oceans15,
title = {DexROV: Enabling Effective Dexterous ROV Operations in Presence of Communication Latency},
author = {J. Gancet and D. Urbina and P. Letier and M. Ilzkovitz and A. Birk and M. Pfingsthorn and P. Weiss and F. Gauch and B. Chemisky and S. Calinon and G. Antonelli and G. Casalino and G. Indiveri and A. Turetta and C. Walen and L. Guilpain},
url = {https://www.researchgate.net/publication/308853452_DexROV_Enabling_effective_dexterous_ROV_operations_in_presence_of_communication_latency},
doi = {10.1109/OCEANS-Genova.2015.7271691},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IEEE Oceans},
abstract = {Subsea interventions in the oil & gas industry as well as in other domains such as archaeology or geological surveys are demanding and costly activities for which robotic solutions are often deployed in addition or in substitution to human divers - contributing to risks and costs cutting. The operation of ROVs (Remotely Operated Vehicles) nevertheless requires significant off-shore dedicated manpower to handle and operate the robotic platform and the supporting vessel. In order to reduce the footprint of operations, DexROV proposes to implement and evaluate novel operation paradigms with safer, more cost effective and time efficient ROV operations. As a keystone of the proposed approach, manned support will in a large extent be delocalized within an onshore ROV control center, possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme also makes provision for advanced dexterous manipulation and semi-autonomous capabilities, leveraging human expertise when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial in the Mediterranean sea.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Subsea interventions in the oil & gas industry as well as in other domains such as archaeology or geological surveys are demanding and costly activities for which robotic solutions are often deployed in addition or in substitution to human divers - contributing to risks and costs cutting. The operation of ROVs (Remotely Operated Vehicles) nevertheless requires significant off-shore dedicated manpower to handle and operate the robotic platform and the supporting vessel. In order to reduce the footprint of operations, DexROV proposes to implement and evaluate novel operation paradigms with safer, more cost effective and time efficient ROV operations. As a keystone of the proposed approach, manned support will in a large extent be delocalized within an onshore ROV control center, possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme also makes provision for advanced dexterous manipulation and semi-autonomous capabilities, leveraging human expertise when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial in the Mediterranean sea.
Gancet, J.; Weiss, P.; Antonelli, G.; Birk, A.; Calinon, S.; Turetta, A.; Walen, C.; Ilzkovitz, M.; Urbina, D.; Letier, P.; Gauch, F.; Indiveri, G.; Casalino, G.; Pfingsthorn, M.; Tanwani, A.; Guilpain, L.
DexROV: Dexterous Undersea Inspection and Maintenance in Presence of Communication Latencies Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2015.
@inproceedings{DexROV-NGCUV15,
title = {DexROV: Dexterous Undersea Inspection and Maintenance in Presence of Communication Latencies},
author = {J. Gancet and P. Weiss and G. Antonelli and A. Birk and S. Calinon and A. Turetta and C. Walen and M. Ilzkovitz and D. Urbina and P. Letier and F. Gauch and G. Indiveri and G. Casalino and M. Pfingsthorn and A. Tanwani and L. Guilpain},
url = {https://www.researchgate.net/publication/283900429_DexROV_Dexterous_Undersea_Inspection_and_Maintenance_in_Presence_of_Communication_Latencies},
doi = {10.1016/j.ifacol.2015.06.036},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {Underwater inspection and maintenance (e.g. in the oil & gas industry) are demanding and costly activities for which ROV based setups are often deployed in addition or in substitution to deep divers - contributing to operations risks and costs cutting. However the operation of a ROV requires significant off-shore dedicated manpower to handle and operate the robotic platform. In order to reduce the burden of operations, DexROV proposes to work out more cost effective and time efficient ROV operations, where manned support is in a large extent delocalized onshore (i.e. from a ROV control center), possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme also makes provision for advanced dexterous manipulation capabilities, exploiting human expertise when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater inspection and maintenance (e.g. in the oil & gas industry) are demanding and costly activities for which ROV based setups are often deployed in addition or in substitution to deep divers - contributing to operations risks and costs cutting. However the operation of a ROV requires significant off-shore dedicated manpower to handle and operate the robotic platform. In order to reduce the burden of operations, DexROV proposes to work out more cost effective and time efficient ROV operations, where manned support is in a large extent delocalized onshore (i.e. from a ROV control center), possibly at a large distance from the actual operations, relying on satellite communications. The proposed scheme also makes provision for advanced dexterous manipulation capabilities, exploiting human expertise when deemed useful. The outcomes of the project will be integrated and evaluated in a series of tests and evaluation campaigns, culminating with a realistic deep sea (1,300 meters) trial.
Köhntopp, Daniel; Lehmann, Benjamin; Kraus, Dieter; Birk, Andreas
Segmentation and Classification Using Active Contours Based Superellipse Fitting on Side Scan Sonar Images for Marine Demining Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2015.
@inproceedings{MarineDemining-ICRA15,
title = {Segmentation and Classification Using Active Contours Based Superellipse Fitting on Side Scan Sonar Images for Marine Demining},
author = {Daniel Köhntopp and Benjamin Lehmann and Dieter Kraus and Andreas Birk},
url = {https://www.researchgate.net/publication/277668547_Segmentation_and_Classification_using_Active_Contours_based_Superellipse_Fitting_on_Side_Scan_Sonar_Images_for_Marine_Demining},
doi = {10.1109/ICRA.2015.7139666},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {This paper proposes a new method for segmenting and classifying seamines on Synthetic Aperture Sonar (SAS) side scan images. The method uses an active contours approach and superellipse a-priori knowledge to segment the image in object, object-shadow and background areas. In contrast to other methods using superellipse constraints, the shape prior is incorporated directly into the segmentation process. This kind of segmentation has the advantage that afterwards the extracted superellipse parameters that describe the object and the object-shadow can directly be used as feature for a classification - this work is hence also of potential interest for general object recognition tasks in other application domains. Several different perspectives of implementing this idea into a suitable algorithm are introduced and compared with each other. Thus, for the evaluation of each method the extracted superellipse features are used for a support vector machine classification. An one against all confusion matrix is generated on a test data set. This result is compared to a related state of the art algorithm. It is shown that our new method is able to correctly classify 170 of 210 objects in a very challenging real world data set and that it yields significant better results than the state of the art comparison.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper proposes a new method for segmenting and classifying seamines on Synthetic Aperture Sonar (SAS) side scan images. The method uses an active contours approach and superellipse a-priori knowledge to segment the image in object, object-shadow and background areas. In contrast to other methods using superellipse constraints, the shape prior is incorporated directly into the segmentation process. This kind of segmentation has the advantage that afterwards the extracted superellipse parameters that describe the object and the object-shadow can directly be used as feature for a classification - this work is hence also of potential interest for general object recognition tasks in other application domains. Several different perspectives of implementing this idea into a suitable algorithm are introduced and compared with each other. Thus, for the evaluation of each method the extracted superellipse features are used for a support vector machine classification. An one against all confusion matrix is generated on a test data set. This result is compared to a related state of the art algorithm. It is shown that our new method is able to correctly classify 170 of 210 objects in a very challenging real world data set and that it yields significant better results than the state of the art comparison.
Kalwa, Joerg; Pascoal, Antonio M.; Ridao, Pere; Glotzbach, Thomas; Brignone, Lorenzo; Bibuli, Marco; Alves, Joao; Birk, Andreas; Silva, Marina Carreiro
EU Project MORPH: Current Status after 3 Years of Cooperation under and above Water Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2015.
@inproceedings{MOPRH-Year3-NGCUV15,
title = {EU Project MORPH: Current Status after 3 Years of Cooperation under and above Water},
author = {Joerg Kalwa and Antonio M. Pascoal and Pere Ridao and Thomas Glotzbach and Lorenzo Brignone and Marco Bibuli and Joao Alves and Andreas Birk and Marina Carreiro Silva},
url = {https://www.researchgate.net/publication/282391786_EU_project_MORPH_Current_Status_After_3_Years_of_Cooperation_Under_and_Above_Water},
doi = {10.1016/j.ifacol.2015.06.019},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {The MORPH project (FP 7, 2012-2016) advances a novel concept of an underwater robotic system composed of a number of spatially separated mobile robot-modules, carrying complementary resources. Instead of being physically coupled, the modules are connected via communication links that allow a limited flow of information among them. Without rigid links, the so called Morph Supra-Vehicle can reconfigure itself and adapt in response to the shape of the terrain, including walls with negative slope.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The MORPH project (FP 7, 2012-2016) advances a novel concept of an underwater robotic system composed of a number of spatially separated mobile robot-modules, carrying complementary resources. Instead of being physically coupled, the modules are connected via communication links that allow a limited flow of information among them. Without rigid links, the so called Morph Supra-Vehicle can reconfigure itself and adapt in response to the shape of the terrain, including walls with negative slope.
Magnusson, Martin; Vaskevicius, Narunas; Stoyanov, Todor; Pathak, Kaustubh; Birk, Andreas
Beyond Points: Evaluating Recent 3D Scan-Matching Algorithms Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2015.
@inproceedings{PlaneReg-NDT-comparison-ICRA15,
title = {Beyond Points: Evaluating Recent 3D Scan-Matching Algorithms},
author = {Martin Magnusson and Narunas Vaskevicius and Todor Stoyanov and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/279911894_Beyond_points_Evaluating_recent_3D_scan-matching_algorithms},
doi = {10.1109/ICRA.2015.7139703},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {Given that 3D scan matching is such a central part of the perception pipeline for robots, thorough and large-scale investigations of scan matching performance are still surprisingly few. A crucial part of the scientific method is to perform experiments that can be replicated by other researchers in order to compare different results. In light of this fact, this paper presents a thorough comparison of 3D scan registration algorithms using a recently published benchmark protocol which makes use of a publicly available challenging data set that covers a wide range of environments. In particular, we evaluate two types of recent 3D registration algorithms - one local and one global. Both approaches take local surface structure into account, rather than matching individual points. After well over 100 000 individual tests, we conclude that algorithms using the normal distributions transform (NDT) provides accurate results compared to a modern implementation of the iterative closest point (ICP) method, when faced with scan data that has little overlap and weak geometric structure. We also demonstrate that the minimally uncertain maximum consensus (MUMC) algorithm provides accurate results in structured environments without needing an initial guess, and that it provides useful measures to detect whether it has succeeded or not. We also propose two amendments to the experimental protocol, in order to provide more valuable results in future implementations.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Given that 3D scan matching is such a central part of the perception pipeline for robots, thorough and large-scale investigations of scan matching performance are still surprisingly few. A crucial part of the scientific method is to perform experiments that can be replicated by other researchers in order to compare different results. In light of this fact, this paper presents a thorough comparison of 3D scan registration algorithms using a recently published benchmark protocol which makes use of a publicly available challenging data set that covers a wide range of environments. In particular, we evaluate two types of recent 3D registration algorithms - one local and one global. Both approaches take local surface structure into account, rather than matching individual points. After well over 100 000 individual tests, we conclude that algorithms using the normal distributions transform (NDT) provides accurate results compared to a modern implementation of the iterative closest point (ICP) method, when faced with scan data that has little overlap and weak geometric structure. We also demonstrate that the minimally uncertain maximum consensus (MUMC) algorithm provides accurate results in structured environments without needing an initial guess, and that it provides useful measures to detect whether it has succeeded or not. We also propose two amendments to the experimental protocol, in order to provide more valuable results in future implementations.
Mihalyi, Razvan-George; Pathak, Kaustubh; Vaskevicius, Narunas; Fromm, Tobias; Birk, Andreas
Robust 3D Object Modeling with a Low-Cost RGBD-Sensor and AR-Markers for Applications with Untrained End-Users Journal Article
In: Robotics and Autonomous Systems (RAS), vol. 66, pp. 1-17, 2015.
@article{3DObjectModeling-ARmarkersUncertainty-RAS15,
title = {Robust 3D Object Modeling with a Low-Cost RGBD-Sensor and AR-Markers for Applications with Untrained End-Users},
author = {Razvan-George Mihalyi and Kaustubh Pathak and Narunas Vaskevicius and Tobias Fromm and Andreas Birk},
url = {https://www.researchgate.net/publication/271274645_Robust_3D_object_modeling_with_a_low-cost_RGBD-sensor_and_AR-markers_for_applications_with_untrained_end-users},
doi = {10.1016/j.robot.2015.01.005},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
journal = {Robotics and Autonomous Systems (RAS)},
volume = {66},
pages = {1-17},
abstract = {An approach for generating textured 3D models of objects without the need for complex infrastructure such as turn-tables or high-end sensors on precisely controlled rails is presented. The method is inexpensive as it uses only a low-cost RGBD sensor, e.g., Microsoft Kinect or ASUS Xtion, and Augmented Reality (AR) markers printed on paper sheets. The sensor can be moved by hand by an untrained person and the AR-markers can be arbitrarily placed in the scene, thus allowing the modeling of objects of a large range of sizes. Due to the use of the simple AR markers, the method is significantly more robust than just using the RGBD sensor or a monocular camera alone and it hence avoids the typical need for manual post-processing of alternative approaches like Kinect-Fusion, 123D Catch, Photosynth, or similar. This article has two main contributions: First, the development of a simple, inexpensive method for the quick and easy digitization of physical objects is presented. Second, the development of an uncertainty model for AR-marker pose estimation is introduced. The latter is of interest beyond the object modeling application presented here. The uncertainty model is used in a graph-based relaxation method to improve model-consistency. Realistic modeling of various objects, such as parcels, sport balls, coffee sacks, human dolls, etc., is experimentally demonstrated. Good model-accuracy is shown for several ground-truth objects with simple geometries and known dimensions. Furthermore, it is shown that the models obtained using the uncertainty model have fewer errors than the ones obtained without it.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An approach for generating textured 3D models of objects without the need for complex infrastructure such as turn-tables or high-end sensors on precisely controlled rails is presented. The method is inexpensive as it uses only a low-cost RGBD sensor, e.g., Microsoft Kinect or ASUS Xtion, and Augmented Reality (AR) markers printed on paper sheets. The sensor can be moved by hand by an untrained person and the AR-markers can be arbitrarily placed in the scene, thus allowing the modeling of objects of a large range of sizes. Due to the use of the simple AR markers, the method is significantly more robust than just using the RGBD sensor or a monocular camera alone and it hence avoids the typical need for manual post-processing of alternative approaches like Kinect-Fusion, 123D Catch, Photosynth, or similar. This article has two main contributions: First, the development of a simple, inexpensive method for the quick and easy digitization of physical objects is presented. Second, the development of an uncertainty model for AR-marker pose estimation is introduced. The latter is of interest beyond the object modeling application presented here. The uncertainty model is used in a graph-based relaxation method to improve model-consistency. Realistic modeling of various objects, such as parcels, sport balls, coffee sacks, human dolls, etc., is experimentally demonstrated. Good model-accuracy is shown for several ground-truth objects with simple geometries and known dimensions. Furthermore, it is shown that the models obtained using the uncertainty model have fewer errors than the ones obtained without it.
Miskovic, Nikola; Pascoal, Antonio; Bibuli, Marco; Caccia, Massimo; Neasham, Jeffrey A.; Birk, Andreas; Egi, Murat; Grammer, Karl; Marroni, Alessandro; Vasilijevic, Antonio; Vukic, Zoran
CADDY Project, Year 1: Overview of Technological Developments and Cooperative Behaviours Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2015.
@inproceedings{CADDY-NGCUV15,
title = {CADDY Project, Year 1: Overview of Technological Developments and Cooperative Behaviours},
author = {Nikola Miskovic and Antonio Pascoal and Marco Bibuli and Massimo Caccia and Jeffrey A. Neasham and Andreas Birk and Murat Egi and Karl Grammer and Alessandro Marroni and Antonio Vasilijevic and Zoran Vukic},
url = {https://www.researchgate.net/publication/281527934_CADDY_Project_Year_1_Overview_of_Technological_Developments_and_Cooperative_Behaviours},
doi = {https://doi.org/10.1016/j.ifacol.2015.06.020},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {"CADDY - Cognitive Autonomous Diving Buddy" is an FP7 project that started in January 2014. Seven partner institutions have joined their efforts towards developing a cognitive underwater robotic system that will help divers during their activities in this hazardous environment. The resulting system will play a threefold role similar to those that a human buddy diver should have: buddy "observer", buddy "slave", and buddy "guide". This paper gives an outline of the CADDY project results during the first year of execution. We focus only on the technical developments and cooperative behaviours that have taken place, in order to keep the overview concise and in line with the workshop topics. Special attention is given to the fleet of new and adapted autonomous marine vehicles used in the project, as well as technologies for perceiving the diver. In addition to that, we give a short overview of cooperative robotic behaviours and present initial results with autonomous surface marine vehicles tracking divers.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
"CADDY - Cognitive Autonomous Diving Buddy" is an FP7 project that started in January 2014. Seven partner institutions have joined their efforts towards developing a cognitive underwater robotic system that will help divers during their activities in this hazardous environment. The resulting system will play a threefold role similar to those that a human buddy diver should have: buddy "observer", buddy "slave", and buddy "guide". This paper gives an outline of the CADDY project results during the first year of execution. We focus only on the technical developments and cooperative behaviours that have taken place, in order to keep the overview concise and in line with the workshop topics. Special attention is given to the fleet of new and adapted autonomous marine vehicles used in the project, as well as technologies for perceiving the diver. In addition to that, we give a short overview of cooperative robotic behaviours and present initial results with autonomous surface marine vehicles tracking divers.
Rathnam, Ravi; Birk, Andreas
Multi-Robot Exploration with AUVs on Cliffs and other 3D Structures with a Predominant Orientation Proceedings Article
In: IEEE Oceans, 2015.
@inproceedings{3Dexploration-cliff-Oceans15,
title = {Multi-Robot Exploration with AUVs on Cliffs and other 3D Structures with a Predominant Orientation},
author = {Ravi Rathnam and Andreas Birk},
url = {https://www.researchgate.net/publication/308840726_Multi-robot_exploration_with_AUVs_on_cliffs_and_other_3D_structures_with_a_predominant_orientation},
doi = {10.1109/OCEANS-Genova.2015.7271563},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IEEE Oceans},
abstract = {A multi-robot exploration algorithm is presented which is targeted at the exploration of 3D structures which predominantly extend along one orientation — the motivating example is underwater cliffs which are locally 3D but that are more globally seen to extend along a continuous vertical surface. The multi-robot exploration aims to maximise the sampling of the environment, causing much denser scans and especially complete scans than a simple lawn mower pattern, i.e., it contains an element of view-planning. This is done by fitting a local plane to the points in the sensor field of view and aligning the robot to view the local plane directly. The distributed multi-robot exploration algorithm is integrated into a complete system consisting of mapping, localisation, and navigation modules. Using a realistic simulation based on real world data from Monte da Guia in the Azores, the algorithm is experimentally validated.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A multi-robot exploration algorithm is presented which is targeted at the exploration of 3D structures which predominantly extend along one orientation — the motivating example is underwater cliffs which are locally 3D but that are more globally seen to extend along a continuous vertical surface. The multi-robot exploration aims to maximise the sampling of the environment, causing much denser scans and especially complete scans than a simple lawn mower pattern, i.e., it contains an element of view-planning. This is done by fitting a local plane to the points in the sensor field of view and aligning the robot to view the local plane directly. The distributed multi-robot exploration algorithm is integrated into a complete system consisting of mapping, localisation, and navigation modules. Using a realistic simulation based on real world data from Monte da Guia in the Azores, the algorithm is experimentally validated.
Schwertfeger, Sören; Birk, Andreas
Map evaluation using matched topology graphs Journal Article
In: Autonomous Robots, vol. 40, no. 5, pp. 761-787, 2015, ISSN: 0929-5593.
@article{MapEval-TopoGraph-AuRo15,
title = {Map evaluation using matched topology graphs},
author = {Sören Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/282391146_Map_evaluation_using_matched_topology_graphs},
doi = {http://dx.doi.org/10.1007/s10514-015-9493-5},
issn = {0929-5593},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
journal = {Autonomous Robots},
volume = {40},
number = {5},
pages = {761-787},
abstract = {Mapping is an important task for mobile robots. The assessment of the quality of maps in a simple, efficient and automated way is not trivial and an ongoing research topic. Here, a new approach for the evaluation of 2D grid maps is presented. This structure-based method makes use of a topology graph, i.e., a topological representation that includes abstracted local metric information. It is shown how the topology graph is constructed from a Voronoi diagram that is pruned and simplified such that only high level topological information remains to concentrate on larger, topologically distinctive places. Several methods for computing the similarity of vertices in two topology graphs, i.e., for performing a place-recognition, are presented. Based on the similarities, it is shown how subgraph-isomorphisms can be efficiently computed and two topology graphs can be matched. The match between the graphs is then used to calculate a number of standard map evaluation attributes like coverage, global accuracy, relative accuracy, consistency, and brokenness. Experiments with robot generated maps are used to highlight the capabilities of the proposed approach and to evaluate the performance of the underlying algorithms.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mapping is an important task for mobile robots. The assessment of the quality of maps in a simple, efficient and automated way is not trivial and an ongoing research topic. Here, a new approach for the evaluation of 2D grid maps is presented. This structure-based method makes use of a topology graph, i.e., a topological representation that includes abstracted local metric information. It is shown how the topology graph is constructed from a Voronoi diagram that is pruned and simplified such that only high level topological information remains to concentrate on larger, topologically distinctive places. Several methods for computing the similarity of vertices in two topology graphs, i.e., for performing a place-recognition, are presented. Based on the similarities, it is shown how subgraph-isomorphisms can be efficiently computed and two topology graphs can be matched. The match between the graphs is then used to calculate a number of standard map evaluation attributes like coverage, global accuracy, relative accuracy, consistency, and brokenness. Experiments with robot generated maps are used to highlight the capabilities of the proposed approach and to evaluate the performance of the underlying algorithms.
Schwertfeger, Sören; Birk, Andreas
Using Fiducials in 3D Map Evaluation Proceedings Article
In: IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR), IEEE Press, 2015.
@inproceedings{MapEvaluation-3Dfiducials-SSRR15,
title = {Using Fiducials in 3D Map Evaluation},
author = {Sören Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/304417495_Using_fiducials_in_3D_map_evaluation},
doi = {10.1109/SSRR.2015.7442997},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {For missions in Safety, Security, and Rescue Robotics (SSRR) maps are often a core deliverable. Hence it is of high interest to assess the quality of maps in a simple and efficient way. Since SSRR is mostly taking place in unstructured environments, 3D mapping has become more and more important. Here a method to evaluate the quality of 3D maps is presented that extends the previously developed 2D Fiducial approach to the third dimension. The artificial features are identified and located in 2D cross-sections of the map as well as in the 3D maps. It is then attempted to proof this concept using a ground truth map and robot generated maps from the RoboCup Rescue competition 2013 in Eindhoven.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
For missions in Safety, Security, and Rescue Robotics (SSRR) maps are often a core deliverable. Hence it is of high interest to assess the quality of maps in a simple and efficient way. Since SSRR is mostly taking place in unstructured environments, 3D mapping has become more and more important. Here a method to evaluate the quality of 3D maps is presented that extends the previously developed 2D Fiducial approach to the third dimension. The artificial features are identified and located in 2D cross-sections of the map as well as in the 3D maps. It is then attempted to proof this concept using a ground truth map and robot generated maps from the RoboCup Rescue competition 2013 in Eindhoven.
2014
Krug, Robert; Stoyanov, Todor; Bonilla, Manuel; Tincani, Vinicio; Vaskevicius, Narunas; Fantoni, Gualtiero; Birk, Andreas; Lilienthal, Achim J.; Bicchi, Antonio
Velvet Fingers: Grasp Planning and Execution for an Underactuated Gripper with Active Surfaces Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2014.
@inproceedings{VelvetGripperJoinedRoblog-ICRA14,
title = {Velvet Fingers: Grasp Planning and Execution for an Underactuated Gripper with Active Surfaces},
author = {Robert Krug and Todor Stoyanov and Manuel Bonilla and Vinicio Tincani and Narunas Vaskevicius and Gualtiero Fantoni and Andreas Birk and Achim J. Lilienthal and Antonio Bicchi},
url = {https://www.researchgate.net/publication/263020588_Velvet_Fingers_Grasp_Planning_and_Execution_for_an_Underactuated_Gripper_with_Active_Surfaces},
doi = {10.1109/ICRA.2014.6907390},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {In this work we tackle the problem of planning grasps for an underactuated gripper which enable it to retrieve target objects from a cluttered environment. Furthermore, we investigate how additional manipulation capabilities of the gripping device, provided by active surfaces on the inside of the fingers, can lead to performance improvement in the grasp execution process. To this end, we employ a simple strategy, in which the target object is 'pulled-in' towards the palm during grasping which results in firm enveloping grasps. We show the effectiveness of the suggested methods by means of experiments conducted in a real-world scenario.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this work we tackle the problem of planning grasps for an underactuated gripper which enable it to retrieve target objects from a cluttered environment. Furthermore, we investigate how additional manipulation capabilities of the gripping device, provided by active surfaces on the inside of the fingers, can lead to performance improvement in the grasp execution process. To this end, we employ a simple strategy, in which the target object is 'pulled-in' towards the palm during grasping which results in firm enveloping grasps. We show the effectiveness of the suggested methods by means of experiments conducted in a real-world scenario.
Mueller, Christian Atanas; Pathak, Kaustubh; Birk, Andreas
Object Shape Categorization in RGBD Images using Hierarchical Graph Constellation Models based on Unsupervisedly Learned Shape Parts described by a Set of Shape Specificity Levels Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014.
@inproceedings{ObjectShapeCategorization-HierarchicalGraphConstellation-IROS14,
title = {Object Shape Categorization in RGBD Images using Hierarchical Graph Constellation Models based on Unsupervisedly Learned Shape Parts described by a Set of Shape Specificity Levels},
author = {Christian Atanas Mueller and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/266086020_Object_Shape_Categorization_in_RGBD_Images_using_Hierarchical_Graph_Constellation_Models_based_on_Unsupervisedly_Learned_Shape_Parts_described_by_a_Set_of_Shape_Specificity_Levels},
doi = {10.1109/IROS.2014.6942984},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
abstract = {We present an approach for object class learning using a part-based shape categorization in RGB-augmented 3D point clouds captured from cluttered indoor scenes with a Kinect-like sensor. We propose an unsupervised hierarchical learning procedure which allows to symbolically classify shape parts by different specificity levels of detailedness of their surface-structural appearance. Further, a hierarchical graphical model is learned that reflects the constellation of classified parts from the set of specificity levels learned in the previous step. Finally an energy minimization inference procedure is applied on the hierarchical graphical model to obtain the corresponding shape category of an object instance consisting of a set of shape parts. Experiments show, due to the proposed classification by learning the hierarchy of shape parts combined with generating the hierarchical graphical constellation model of shape parts, the additional evidence on different levels of shape detailedness is a major factor that leads to a more robust and accurate categorization compared to a flat approach. The experiments are conducted for shape categories (sack, barrel and parcel) which typically appear in visuoperceptual-challenging scenarios of logistic processes.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present an approach for object class learning using a part-based shape categorization in RGB-augmented 3D point clouds captured from cluttered indoor scenes with a Kinect-like sensor. We propose an unsupervised hierarchical learning procedure which allows to symbolically classify shape parts by different specificity levels of detailedness of their surface-structural appearance. Further, a hierarchical graphical model is learned that reflects the constellation of classified parts from the set of specificity levels learned in the previous step. Finally an energy minimization inference procedure is applied on the hierarchical graphical model to obtain the corresponding shape category of an object instance consisting of a set of shape parts. Experiments show, due to the proposed classification by learning the hierarchy of shape parts combined with generating the hierarchical graphical constellation model of shape parts, the additional evidence on different levels of shape detailedness is a major factor that leads to a more robust and accurate categorization compared to a flat approach. The experiments are conducted for shape categories (sack, barrel and parcel) which typically appear in visuoperceptual-challenging scenarios of logistic processes.
Pfingsthorn, Max; Birk, Andreas
Representing and Solving Local and Global Ambiguities as Multimodal and Hyperedge Constraints in a Generalized Graph SLAM Framework Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2014.
@inproceedings{GeneralizedGraphSLAM-BremenData-ICRA14,
title = {Representing and Solving Local and Global Ambiguities as Multimodal and Hyperedge Constraints in a Generalized Graph SLAM Framework},
author = {Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/277007022_Representing_and_solving_local_and_global_ambiguities_as_multimodal_and_hyperedge_constraints_in_a_generalized_graph_SLAM_framework},
doi = {10.1109/ICRA.2014.6907481},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {Graph-based Simultaneous Localization and Mapping (SLAM) has experienced a recent surge towards robust methods. These methods take the combinatorial aspect of data association into account by allowing decisions of the graph topology to be made during optimization. In this paper, the Generalized Graph SLAM framework for SLAM under ambiguous data association is presented, and a formal description of using hyperedges to encode uncertain loop closures is given for the first time. The framework combines both hyperedges and multimodal Mixture of Gaussian constraints to cover all sources of ambiguity in SLAM. An extension of the authors' multimodal Prefilter method is developed to find good initial conditions in such a generalized multimodal hypergraph. Experiments on a real world 3D dataset show that the novel multimodal hypergraph Prefilter method is both significantly more robust and faster than other robust state-of-the-art methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Graph-based Simultaneous Localization and Mapping (SLAM) has experienced a recent surge towards robust methods. These methods take the combinatorial aspect of data association into account by allowing decisions of the graph topology to be made during optimization. In this paper, the Generalized Graph SLAM framework for SLAM under ambiguous data association is presented, and a formal description of using hyperedges to encode uncertain loop closures is given for the first time. The framework combines both hyperedges and multimodal Mixture of Gaussian constraints to cover all sources of ambiguity in SLAM. An extension of the authors' multimodal Prefilter method is developed to find good initial conditions in such a generalized multimodal hypergraph. Experiments on a real world 3D dataset show that the novel multimodal hypergraph Prefilter method is both significantly more robust and faster than other robust state-of-the-art methods.
Pfingsthorn, Max; Birk, Andreas; Ferreira, Fausto; Veruggio, Gianmarco; Caccia, Massimo; Bruzzone, Gabriele
Large-Scale Image Mosaicking using Multimodal Hyperedge Constraints from Multiple Registration Methods within the Generalized Graph SLAM Framework Proceedings Article
In: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2014.
@inproceedings{GeneralizedGraphSLAM-UnderwaterMosaicking-IROS14,
title = {Large-Scale Image Mosaicking using Multimodal Hyperedge Constraints from Multiple Registration Methods within the Generalized Graph SLAM Framework},
author = {Max Pfingsthorn and Andreas Birk and Fausto Ferreira and Gianmarco Veruggio and Massimo Caccia and Gabriele Bruzzone},
url = {https://www.researchgate.net/publication/287510162_Large-scale_image_mosaicking_using_multimodal_hyperedge_constraints_from_multiple_registration_methods_within_the_Generalized_Graph_SLAM_framework},
doi = {10.1109/IROS.2014.6943209},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
abstract = {Underwater image mosaicking is an important tool for visual surveys, object detection, and as a means to control the underwater robot if done online. Such application areas can benefit significantly from a recent focus on robust methods for graph-based Simultaneous Localization and Mapping (SLAM). This paper focuses on two contributions: An approach to combine registration results from multiple methods in multimodal constraints and, up to the authors' knowledge, the first method to generate hyperedge constraints from state-of-the-art place recognition techniques. Both contributions are implemented within the Generalized Graph SLAM framework. Experimental results show that the methods generate informative constraints and that the authors' Prefilter method outperforms related methods on a large underwater image dataset processed with these methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater image mosaicking is an important tool for visual surveys, object detection, and as a means to control the underwater robot if done online. Such application areas can benefit significantly from a recent focus on robust methods for graph-based Simultaneous Localization and Mapping (SLAM). This paper focuses on two contributions: An approach to combine registration results from multiple methods in multimodal constraints and, up to the authors' knowledge, the first method to generate hyperedge constraints from state-of-the-art place recognition techniques. Both contributions are implemented within the Generalized Graph SLAM framework. Experimental results show that the methods generate informative constraints and that the authors' Prefilter method outperforms related methods on a large underwater image dataset processed with these methods.
Vaskevicius, Narunas; Mueller, Christian Atanas; Bonilla, Manuel; Tincani, Vinicio; Stoyanov, Todor; Fantoni, Gualtiero; Pathak, Kaustubh; Lilienthal, Achim J.; Bicchi, Antonio; Birk, Andreas
Object Recognition and Localization for Robust Grasping with a Dexterous Gripper in the Context of Container Unloading Proceedings Article
In: IEEE International Conference on Automation Science and Engineering (CASE), 2014.
@inproceedings{ObjectRecognitionVelvetGripper-CASE14,
title = {Object Recognition and Localization for Robust Grasping with a Dexterous Gripper in the Context of Container Unloading},
author = {Narunas Vaskevicius and Christian Atanas Mueller and Manuel Bonilla and Vinicio Tincani and Todor Stoyanov and Gualtiero Fantoni and Kaustubh Pathak and Achim J. Lilienthal and Antonio Bicchi and Andreas Birk},
url = {https://www.researchgate.net/publication/286723602_Object_recognition_and_localization_for_robust_grasping_with_a_dexterous_gripper_in_the_context_of_container_unloading},
doi = {10.1109/CoASE.2014.6899490},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE International Conference on Automation Science and Engineering (CASE)},
abstract = {The work presented here is embedded in research on an industrial application scenario, namely autonomous shipping-container unloading, which has several challenging constraints: the scene is very cluttered, objects can be much larger than in common table-top scenarios; the perception must be highly robust, while being as fast as possible. These contradicting goals force a compromise between speed and accuracy. In this work, we investigate a state of the art perception system integrated with a dexterous gripper. In particular, we are interested in pose estimation errors from the recognition module and whether these errors can be handled by the abilities of the gripper.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The work presented here is embedded in research on an industrial application scenario, namely autonomous shipping-container unloading, which has several challenging constraints: the scene is very cluttered, objects can be much larger than in common table-top scenarios; the perception must be highly robust, while being as fast as possible. These contradicting goals force a compromise between speed and accuracy. In this work, we investigate a state of the art perception system integrated with a dexterous gripper. In particular, we are interested in pose estimation errors from the recognition module and whether these errors can be handled by the abilities of the gripper.
Vaskevicius, Narunas; Pathak, Kaustubh; Birk, Andreas
Fitting Superquadrics in Noisy, Partial Views from a Low-cost RGBD Sensor for Recognition and Localization of Sacks in Autonomous Unloading of Shipping Containers Proceedings Article
In: IEEE International Conference on Automation Science and Engineering (CASE), 2014.
@inproceedings{SuperquadricsSacksRecognition-CASE14,
title = {Fitting Superquadrics in Noisy, Partial Views from a Low-cost RGBD Sensor for Recognition and Localization of Sacks in Autonomous Unloading of Shipping Containers},
author = {Narunas Vaskevicius and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/286705157_Fitting_superquadrics_in_noisy_partial_views_from_a_low-cost_RGBD_sensor_for_recognition_and_localization_of_sacks_in_autonomous_unloading_of_shipping_containers},
doi = {10.1109/CoASE.2014.6899335},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {IEEE International Conference on Automation Science and Engineering (CASE)},
abstract = {A significant amount of cargo is transported in sacks, e.g., coffee and cacao, which are predominantly handled through manual labor, e.g., when being unloaded from shipping containers. There is hence a huge potential for automation. We present here a perception pipeline to recognize and localize sacks with a low-cost sensor. The pipeline is embedded in an industrial demonstration system for container unloading. In addition to the application background, there are two main contributions presented in this paper. First, we introduce a new numerically stable form of superquadric fitting. This is of interest for the application of superquadrics in general far beyond the concrete application scenario in this paper. Second, we introduce a fast convexity test between two neighboring patches that leads to a robust segmentation for sack/bag-recognition.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A significant amount of cargo is transported in sacks, e.g., coffee and cacao, which are predominantly handled through manual labor, e.g., when being unloaded from shipping containers. There is hence a huge potential for automation. We present here a perception pipeline to recognize and localize sacks with a low-cost sensor. The pipeline is embedded in an industrial demonstration system for container unloading. In addition to the application background, there are two main contributions presented in this paper. First, we introduce a new numerically stable form of superquadric fitting. This is of interest for the application of superquadrics in general far beyond the concrete application scenario in this paper. Second, we introduce a fast convexity test between two neighboring patches that leads to a robust segmentation for sack/bag-recognition.
2013
Bülow, Heiko; Birk, Andreas
Spectral 6-DOF Registration of Noisy 3D Range Data with Partial Overlap Journal Article
In: IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 35, no. 4, pp. 954-969, 2013.
@article{SRMR-6DoF-registration-PAMI12,
title = {Spectral 6-DOF Registration of Noisy 3D Range Data with Partial Overlap},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/230623021_Spectral_6DOF_Registration_of_Noisy_3D_Range_Data_with_Partial_Overlap},
doi = {10.1109/TPAMI.2012.173},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI)},
volume = {35},
number = {4},
pages = {954-969},
abstract = {We present Spectral Registration with Multilayer Resampling (SRMR) as a 6 Degrees Of Freedom (DOF) registration method for noisy 3D data with partial overlap. The algorithm is based on decoupling 3D rotation from 3D translation by a corresponding resampling process of the spectral magnitude of a 3D Fast Fourier Transform (FFT) calculation on discretized 3D range data. The registration of all 6DOF is then subsequently carried out with spectral registrations using Phase Only Matched Filtering (POMF). There are two main aspects for the fast and robust registration of Euler angles from spherical information in SRMR. First of all, there is the permanent use of phase matching. Second, based on the FFT on a discrete Cartesian grid, not only one spherical layer but a complete stack of layers is processed in one step. Experiments are presented with challenging data sets with respect to interference and overlap. The results include the fast and robust registration of artificially transformed data for ground truth comparison, scans from the Stanford Bunny data-set, high end 3D laser range finder (LRF) scans of a city center, and range data from a low-cost actuated LRF in a disaster response scenario.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We present Spectral Registration with Multilayer Resampling (SRMR) as a 6 Degrees Of Freedom (DOF) registration method for noisy 3D data with partial overlap. The algorithm is based on decoupling 3D rotation from 3D translation by a corresponding resampling process of the spectral magnitude of a 3D Fast Fourier Transform (FFT) calculation on discretized 3D range data. The registration of all 6DOF is then subsequently carried out with spectral registrations using Phase Only Matched Filtering (POMF). There are two main aspects for the fast and robust registration of Euler angles from spherical information in SRMR. First of all, there is the permanent use of phase matching. Second, based on the FFT on a discrete Cartesian grid, not only one spherical layer but a complete stack of layers is processed in one step. Experiments are presented with challenging data sets with respect to interference and overlap. The results include the fast and robust registration of artificially transformed data for ground truth comparison, scans from the Stanford Bunny data-set, high end 3D laser range finder (LRF) scans of a city center, and range data from a low-cost actuated LRF in a disaster response scenario.
Bülow, Heiko; Sokolovski, Igor; Pfingsthorn, Max; Birk, Andreas
Underwater Stereo Data Acquisition and 3D Registration with a Spectral Method Proceedings Article
In: IEEE Oceans, IEEE Press, 2013.
@inproceedings{UnderwaterStereo-Oceans13,
title = {Underwater Stereo Data Acquisition and 3D Registration with a Spectral Method},
author = {Heiko Bülow and Igor Sokolovski and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/261024933_Underwater_stereo_data_acquisition_and_3D_registration_with_a_spectral_method},
doi = {10.1109/OCEANS-Bergen.2013.6608115},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE Press},
abstract = {An underwater stereo camera system is presented that is an initial development towards an online 3D mapping system, i.e., that can generate 3D maps of underwater sites in real time during the mission. The system consists of according hardware in form of a stereo camera plus an onboard computer in an underwater housing plus a software side for the processing. A core software component in the system is Spectral Registration with Multilayer Resampling (SRMR) as a 6 Degrees Of Freedom (DOF) registration method for noisy 3D data with partial overlap. Initial results from a first field test off the island of Faial in the Azores are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An underwater stereo camera system is presented that is an initial development towards an online 3D mapping system, i.e., that can generate 3D maps of underwater sites in real time during the mission. The system consists of according hardware in form of a stereo camera plus an onboard computer in an underwater housing plus a software side for the processing. A core software component in the system is Spectral Registration with Multilayer Resampling (SRMR) as a 6 Degrees Of Freedom (DOF) registration method for noisy 3D data with partial overlap. Initial results from a first field test off the island of Faial in the Azores are presented.
Matsunoa, Fumitoshi; Noda, Itsuki; Murphy, Robin; Birk, Andreas; (eds),
Spec. Iss. on Safety, Security, and Rescue Robotics Journal Article
In: Advanced Robotics, vol. 27, no. 5, 2013.
@article{SpecIss-SSRR-AdvRob13,
title = {Spec. Iss. on Safety, Security, and Rescue Robotics},
author = {Fumitoshi Matsunoa and Itsuki Noda and Robin Murphy and Andreas Birk and (eds)},
doi = {10.1080/01691864.2013.774511},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
journal = {Advanced Robotics},
volume = {27},
number = {5},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mihalyi, Razvan; Pathak, Kaustubh; Vaskevicius, Narunas; Birk, Andreas
Uncertainty Estimation of AR-Marker Poses for Graph-SLAM Optimization in 3D Object Model Generation with RGBD Data Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), 2013.
@inproceedings{RoblogModelingTool-IROS2013,
title = {Uncertainty Estimation of AR-Marker Poses for Graph-SLAM Optimization in 3D Object Model Generation with RGBD Data},
author = {Razvan Mihalyi and Kaustubh Pathak and Narunas Vaskevicius and Andreas Birk},
url = {https://www.researchgate.net/publication/261384924_Uncertainty_estimation_of_AR-marker_poses_for_graph-SLAM_optimization_in_3D_object_model_generation_with_RGBD_data},
doi = {10.1109/IROS.2013.6696594},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
abstract = {This paper presents an approach to acquire textured 3D models of objects without the need for sophisticated hardware infrastructures. The approach is inexpensive, using a low-cost Microsoft Kinect RGB-D sensor and Augmented Reality (AR) markers printed on paper sheets. The AR-markers can be freely placed in the scene, allowing the modeling of objects of various sizes, and the sensor can be moved by the hand of an untrained person. To generate usable models with this very inexpensive and simple setup, the sequence of RGB-D scans is embedded in a graph-based optimizer for automatic post-refinement. The main novelty of this contribution is the development of an uncertainty model for an AR-marker. The AR-marker uncertainty models are used as constraints in an optimization problem to better estimate the object pose. The models are in the end further fine-tuned by a standard point-based registration algorithm. The results section presents realistic models of various objects generated using this system, e.g., parcels, sport balls, human dolls etc. Additionally, a quantitative analysis is presented using objects of known dimensions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents an approach to acquire textured 3D models of objects without the need for sophisticated hardware infrastructures. The approach is inexpensive, using a low-cost Microsoft Kinect RGB-D sensor and Augmented Reality (AR) markers printed on paper sheets. The AR-markers can be freely placed in the scene, allowing the modeling of objects of various sizes, and the sensor can be moved by the hand of an untrained person. To generate usable models with this very inexpensive and simple setup, the sequence of RGB-D scans is embedded in a graph-based optimizer for automatic post-refinement. The main novelty of this contribution is the development of an uncertainty model for an AR-marker. The AR-marker uncertainty models are used as constraints in an optimization problem to better estimate the object pose. The models are in the end further fine-tuned by a standard point-based registration algorithm. The results section presents realistic models of various objects generated using this system, e.g., parcels, sport balls, human dolls etc. Additionally, a quantitative analysis is presented using objects of known dimensions.
Mueller, Christian Atanas; Pathak, Kaustubh; Birk, Andreas
Object Recognition in RGBD Images of Cluttered Environments using Graph-Based Categorization with Unsupervised Learning of Shape Parts Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), 2013.
@inproceedings{NeuralGas-ObjectRecognition-IROS2013,
title = {Object Recognition in RGBD Images of Cluttered Environments using Graph-Based Categorization with Unsupervised Learning of Shape Parts},
author = {Christian Atanas Mueller and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/258757922_Object_Recognition_in_RGBD_Images_of_Cluttered_Environments_Using_Graph-Based_Categorization_with_Unsupervised_Learning_of_Shape_Parts},
doi = {10.1109/IROS.2013.6696671},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
abstract = {We present an approach for object class learning using a part-based shape categorization in RGB-augmented 3D point clouds captured from cluttered indoor scenes with a Kinect-like sensor. A graph representation is used to detect and categorize object instances based on part-constellations found in scenes. No assumptions like objects being placed on planar surfaces or constraints on their poses are required. Our approach consists of the following steps: 1) a Mean- Shift-based over-segmentation of a point cloud into atomic patches; 2) use of topological and geometric features to merge surface-homogeneous atomic patches into super patches; 3) an unsupervised classification of these parts that allows to symbolically label distinctively unknown object parts by their surface-structural appearance; and finally, 4) a graph generation procedure that reflects the constellation of the detected parts from object instances of certain shape categories. Furthermore, an inference procedure is presented that processes extracted part constellations of a scene to detect and categorize object instances. Experiments with challenging, cluttered scenes show that the segmentation procedure provides salient parts of objects which lead to a good categorization performance using the graph-based constellation model concept.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We present an approach for object class learning using a part-based shape categorization in RGB-augmented 3D point clouds captured from cluttered indoor scenes with a Kinect-like sensor. A graph representation is used to detect and categorize object instances based on part-constellations found in scenes. No assumptions like objects being placed on planar surfaces or constraints on their poses are required. Our approach consists of the following steps: 1) a Mean- Shift-based over-segmentation of a point cloud into atomic patches; 2) use of topological and geometric features to merge surface-homogeneous atomic patches into super patches; 3) an unsupervised classification of these parts that allows to symbolically label distinctively unknown object parts by their surface-structural appearance; and finally, 4) a graph generation procedure that reflects the constellation of the detected parts from object instances of certain shape categories. Furthermore, an inference procedure is presented that processes extracted part constellations of a scene to detect and categorize object instances. Experiments with challenging, cluttered scenes show that the segmentation procedure provides salient parts of objects which lead to a good categorization performance using the graph-based constellation model concept.
Pathak, Kaustubh; Pfingsthorn, Max; Bülow, Heiko; Birk, Andreas
Robust Estimation of Camera-Tilt for iFMI based Photo-Mapping using a Calibrated Monocular Camera Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2013.
@inproceedings{IFMI-tilt-ICRA13,
title = {Robust Estimation of Camera-Tilt for iFMI based Photo-Mapping using a Calibrated Monocular Camera},
author = {Kaustubh Pathak and Max Pfingsthorn and Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/261506315_Robust_estimation_of_camera-tilt_for_iFMI_based_underwater_photo-mapping_using_a_calibrated_monocular_camera},
doi = {10.1109/ICRA.2013.6631307},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {An autonomous aerial or underwater robot can create photo-maps using a downward looking camera to not only compute its odometry visually but also to provide a useful and intuitively understandable representation of the environment explored by it. The improved Fourier Mellin Invariant (iFMI) registration is a spectral registration method, which has specific benefits, especially high robustness in featureless scenarios, but it only allows registrations of 2D translations, rotation, and scale. The method is extended here to incorporate tilt using parallax information. To this end, we extend the well-known four-point algorithm for planar homography. We show that using the decomposition of the planar homography to compute the tilt is very noise-prone, and propose a way of increasing this accuracy based on a parallax to noise metric. Although our general approach can be used with local scale invariant image features, we implement the tilt-correction based on an extension of our frequency-based approach to determine the image motion-field. Two experiments are presented to show the efficacy and applicability of our approach: An analysis of a simulated data set with ground truth is used to quantify the robustness of our novel method relative to the same four-point method using only SIFT (Scale Invariant Feature Transform) features. A second data set is used to present similar results with real-world data.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An autonomous aerial or underwater robot can create photo-maps using a downward looking camera to not only compute its odometry visually but also to provide a useful and intuitively understandable representation of the environment explored by it. The improved Fourier Mellin Invariant (iFMI) registration is a spectral registration method, which has specific benefits, especially high robustness in featureless scenarios, but it only allows registrations of 2D translations, rotation, and scale. The method is extended here to incorporate tilt using parallax information. To this end, we extend the well-known four-point algorithm for planar homography. We show that using the decomposition of the planar homography to compute the tilt is very noise-prone, and propose a way of increasing this accuracy based on a parallax to noise metric. Although our general approach can be used with local scale invariant image features, we implement the tilt-correction based on an extension of our frequency-based approach to determine the image motion-field. Two experiments are presented to show the efficacy and applicability of our approach: An analysis of a simulated data set with ground truth is used to quantify the robustness of our novel method relative to the same four-point method using only SIFT (Scale Invariant Feature Transform) features. A second data set is used to present similar results with real-world data.
Pfingsthorn, M.; Bülow, H.; Birk, A.; Ferreira, F.; Veruggio, G.; Caccia, M.; Bruzzone, G.
Large-Scale Mosaicking with Spectral Registration based Simultaneous Localization and Mapping (iFMI-SLAM) in the Ligurian Sea Proceedings Article
In: IEEE Oceans, IEEE Press, 2013.
@inproceedings{IFMI-Mosaicking-LigurianSea-Oceans13,
title = {Large-Scale Mosaicking with Spectral Registration based Simultaneous Localization and Mapping (iFMI-SLAM) in the Ligurian Sea},
author = {M. Pfingsthorn and H. Bülow and A. Birk and F. Ferreira and G. Veruggio and M. Caccia and G. Bruzzone},
url = {https://www.researchgate.net/publication/287510162_Large-scale_image_mosaicking_using_multimodal_hyperedge_constraints_from_multiple_registration_methods_within_the_Generalized_Graph_SLAM_framework},
doi = {10.1109/IROS.2014.6943209},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE Press},
abstract = {Underwater image mosaicking is an important tool for visual surveys, object detection, and as a means to control the underwater robot if done online. Such application areas can benefit significantly from a recent focus on robust methods for graph-based Simultaneous Localization and Mapping (SLAM). This paper focuses on two contributions: An approach to combine registration results from multiple methods in multimodal constraints and, up to the authors' knowledge, the first method to generate hyperedge constraints from state-of-the-art place recognition techniques. Both contributions are implemented within the Generalized Graph SLAM framework. Experimental results show that the methods generate informative constraints and that the authors' Prefilter method outperforms related methods on a large underwater image dataset processed with these methods.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Underwater image mosaicking is an important tool for visual surveys, object detection, and as a means to control the underwater robot if done online. Such application areas can benefit significantly from a recent focus on robust methods for graph-based Simultaneous Localization and Mapping (SLAM). This paper focuses on two contributions: An approach to combine registration results from multiple methods in multimodal constraints and, up to the authors' knowledge, the first method to generate hyperedge constraints from state-of-the-art place recognition techniques. Both contributions are implemented within the Generalized Graph SLAM framework. Experimental results show that the methods generate informative constraints and that the authors' Prefilter method outperforms related methods on a large underwater image dataset processed with these methods.
Pfingsthorn, Max; Birk, Andreas
Simultaneous Localization and Mapping (SLAM) with Multimodal Probability Distributions Journal Article
In: The international Journal of Robotics Research, vol. 32, no. 2, pp. 143-171, 2013.
@article{MultimodalGraphSLAM-IJRR12,
title = {Simultaneous Localization and Mapping (SLAM) with Multimodal Probability Distributions},
author = {Max Pfingsthorn and Andreas Birk},
url = {http://ijr.sagepub.com/content/early/2012/10/05/0278364912461540.abstract},
doi = {10.1177/0278364912461540},
year = {2013},
date = {2013-01-01},
journal = {The international Journal of Robotics Research},
volume = {32},
number = {2},
pages = {143-171},
abstract = {Simultaneous Localization and Mapping (SLAM) has focused on noisy but unique data associations resulting in linear Gaussian uncertainty models. However, a unique decision is often not possible using only local information, giving rise to ambiguities that have to be resolved globally during optimization. To solve this problem, the pose graph data structure is extended here by multimodal constraints modeled by mixtures of Gaussians (MoG). Furthermore, optimization methods for this novel formulation are introduced, namely (a) robust iteratively reweighted least squares, and (b) Prefilter Stochastic Gradient Descent (SGD) where a preprocessing step determines globally consistent modes before applying SGD. In addition, a variant of the Prefilter method (b) is introduced in form of (c) Prefilter Levenberg–Marquardt. The methods are compared with traditional state-of-the-art optimization methods including (d) Stochastic Gradient Descent and (e) Levenberg–Marquardt as well as (f) Particle filter SLAM and with (g) an optimal exhaustive algorithm. Experiments show that ambiguities significantly impact state-of-the-art methods, and that the novel Prefilter methods (b) and (c) perform best. This is further substantiated with experiments using real-world data. To this end, a method to generate MoG constraints from a plane-based registration algorithm is introduced and used for 3D SLAM under ambiguities.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Simultaneous Localization and Mapping (SLAM) has focused on noisy but unique data associations resulting in linear Gaussian uncertainty models. However, a unique decision is often not possible using only local information, giving rise to ambiguities that have to be resolved globally during optimization. To solve this problem, the pose graph data structure is extended here by multimodal constraints modeled by mixtures of Gaussians (MoG). Furthermore, optimization methods for this novel formulation are introduced, namely (a) robust iteratively reweighted least squares, and (b) Prefilter Stochastic Gradient Descent (SGD) where a preprocessing step determines globally consistent modes before applying SGD. In addition, a variant of the Prefilter method (b) is introduced in form of (c) Prefilter Levenberg–Marquardt. The methods are compared with traditional state-of-the-art optimization methods including (d) Stochastic Gradient Descent and (e) Levenberg–Marquardt as well as (f) Particle filter SLAM and with (g) an optimal exhaustive algorithm. Experiments show that ambiguities significantly impact state-of-the-art methods, and that the novel Prefilter methods (b) and (c) perform best. This is further substantiated with experiments using real-world data. To this end, a method to generate MoG constraints from a plane-based registration algorithm is introduced and used for 3D SLAM under ambiguities.
Rathnam, Ravi; Birk, Andreas
Initial Results of Cooperative AUV Exploration in a High-Fidelity Simulation using Real-World Data from Monte da Guia, Azores Proceedings Article
In: IEEE Oceans, IEEE Press, 2013.
@inproceedings{CooperativeExploration-MonteDaGuia-Oceans13,
title = {Initial Results of Cooperative AUV Exploration in a High-Fidelity Simulation using Real-World Data from Monte da Guia, Azores},
author = {Ravi Rathnam and Andreas Birk},
url = {https://www.researchgate.net/publication/260292079_Initial_Results_of_Cooperative_AUV_Exploration_in_a_High-Fidelity_Simulation_using_Real-World_Data_from_Monte_da_Guia_Azores},
doi = {10.1109/OCEANS-Bergen.2013.6608147},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE Press},
abstract = {An algorithm for distributed exploration in 3D is presented which is integrated into a complete control infrastructure for Autonomous Underwater Vehicles (AUVs) containing sensors, mapping, navigation, and control of actuators. The robots cooperate to produce a common map, which is then used to explore the environment. In order to ensure the building of the common map, the algorithm always keeps the robots within communication range of each other. The method is based on a greedy optimization strategy which uses a heuristic utility function which makes it computationally efficient but it can also lead to local minimums. However, related deadlocks can be easily detected during the exploration process and there is an efficient strategy to recover from them. An environment based on real world data is created in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. The effect of the sensor range and the number of robots on the algorithm is investigated.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An algorithm for distributed exploration in 3D is presented which is integrated into a complete control infrastructure for Autonomous Underwater Vehicles (AUVs) containing sensors, mapping, navigation, and control of actuators. The robots cooperate to produce a common map, which is then used to explore the environment. In order to ensure the building of the common map, the algorithm always keeps the robots within communication range of each other. The method is based on a greedy optimization strategy which uses a heuristic utility function which makes it computationally efficient but it can also lead to local minimums. However, related deadlocks can be easily detected during the exploration process and there is an efficient strategy to recover from them. An environment based on real world data is created in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. The effect of the sensor range and the number of robots on the algorithm is investigated.
Rathnam, Ravi Kulan; Birk, Andreas
A Distributed Algorithm for Cooperative 3D Exploration under Communication Constraints Journal Article
In: Paladyn, Journal of Behavioral Robotics, vol. 4, no. 4, pp. 223–232, 2013.
@article{CoordExpl3D-JBR13,
title = {A Distributed Algorithm for Cooperative 3D Exploration under Communication Constraints},
author = {Ravi Kulan Rathnam and Andreas Birk},
url = {https://www.degruyter.com/document/doi/10.2478/pjbr-2013-0020/html},
doi = {10.2478/pjbr-2013-0020},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
journal = {Paladyn, Journal of Behavioral Robotics},
volume = {4},
number = {4},
pages = {223–232},
abstract = {An algorithm for distributed exploration in 3D is presented which always keeps the robots within communication range of each other. The method is based on a greedy optimization strategy that uses a heuristic utility function. This makes it computationally very efficient but it can also lead to local minimums; but related deadlocks can be easily detected during the exploration process and there is an efficient strategy to recover from them. The exploration algorithm is integrated into a complete control infrastructure for Autonomous Underwater Vehicles (AUV) containing sensors, mapping, navigation, and control of actuators. The algorithm is tested in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. The effect of the communication range and the number of robots on the algorithm is investigated.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An algorithm for distributed exploration in 3D is presented which always keeps the robots within communication range of each other. The method is based on a greedy optimization strategy that uses a heuristic utility function. This makes it computationally very efficient but it can also lead to local minimums; but related deadlocks can be easily detected during the exploration process and there is an efficient strategy to recover from them. The exploration algorithm is integrated into a complete control infrastructure for Autonomous Underwater Vehicles (AUV) containing sensors, mapping, navigation, and control of actuators. The algorithm is tested in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. The effect of the communication range and the number of robots on the algorithm is investigated.
Schwertfeger, Sören; Birk, Andreas
Evaluation of Map Quality by Matching and Scoring High-Level, Topological Map Structures Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2013.
@inproceedings{MapEvaluation-TopologicalStructures-ICRA13,
title = {Evaluation of Map Quality by Matching and Scoring High-Level, Topological Map Structures},
author = {Sören Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/261416033_Evaluation_of_map_quality_by_matching_and_scoring_high-level_topological_map_structures10.1109/ICRA.2013.6630876},
doi = {10.1109/ICRA.2013.6630876},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {Mapping is an important task for mobile robots. But assessing the quality of maps in a simple, efficient and automated way is not trivial and an ongoing research topic. A new approach on map evaluation is presented here. It is based on Topology Graphs as a topological, abstracted representation of 2D grid maps. The Topology Graphs are derived from Voronoi Diagrams that get post-processed to capture the high-level spatial structures. Based on a similarity metric on vertices in Topology Graphs, the vertices can be matched across maps and spatial (dis)similarities and hence errors in the mapping can be identified and measured. More precisely, the vertex-similarity is the basis to match the structures of Topology Graphs up to the identification of subgraph isomorphisms through wave-front propagation. This allows to determine important map quality attributes up to very challenging structural elements like brokenness, i.e., the number of locally correct partitions in the candidate map and their relative placement towards each other. Experiments with real robot generated maps including examples from various teams in the RoboCup Rescue competition are used to validate the usefulness of this method for map quality assessment.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mapping is an important task for mobile robots. But assessing the quality of maps in a simple, efficient and automated way is not trivial and an ongoing research topic. A new approach on map evaluation is presented here. It is based on Topology Graphs as a topological, abstracted representation of 2D grid maps. The Topology Graphs are derived from Voronoi Diagrams that get post-processed to capture the high-level spatial structures. Based on a similarity metric on vertices in Topology Graphs, the vertices can be matched across maps and spatial (dis)similarities and hence errors in the mapping can be identified and measured. More precisely, the vertex-similarity is the basis to match the structures of Topology Graphs up to the identification of subgraph isomorphisms through wave-front propagation. This allows to determine important map quality attributes up to very challenging structural elements like brokenness, i.e., the number of locally correct partitions in the candidate map and their relative placement towards each other. Experiments with real robot generated maps including examples from various teams in the RoboCup Rescue competition are used to validate the usefulness of this method for map quality assessment.
Kalwa, J.; Carreiro-Silva, M.; Tempera, F.; Fontes, J.; Santos, R. S.; Fabri, M. C.; Brignone, L.; Ridao, P.; Birk, A.; Glotzbach, T.; Caccia, M.; Alves, J.; Pascoal, A.
The MORPH concept and its application in marine research Proceedings Article
In: MTS/IEEE OCEANS, pp. 1-8, 2013.
@inproceedings{MORPH-Oceans13,
title = {The MORPH concept and its application in marine research},
author = {J. Kalwa and M. Carreiro-Silva and F. Tempera and J. Fontes and R. S. Santos and M. C. Fabri and L. Brignone and P. Ridao and A. Birk and T. Glotzbach and M. Caccia and J. Alves and A. Pascoal},
url = {https://www.researchgate.net/publication/256981298_The_MORPH_Concept_and_Its_Application_in_Marine_Research},
doi = {10.1109/OCEANS-Bergen.2013.6607988},
year = {2013},
date = {2013-01-01},
urldate = {2013-01-01},
booktitle = {MTS/IEEE OCEANS},
pages = {1-8},
abstract = {The MORPH project (FP 7, 2012-2016) aims at the development of efficient methods and tools to map the underwater environment in situations that defy existing technology. Namely, missions that involve underwater surveying and marine habitat mapping of rugged terrain and structures with full 3D complexity, including near-vertical cliffs. Potential applications include the study of cold water coral reefs or ecosystems in underwater canyons. For mapping purposes, the project advances the novel concept of an underwater robotic sensor carrier consisting of a number of spatially separated mobile robotic modules, each of them carrying complementary sensors. The modules are connected virtually via wireless communication links. Free from the constraints of rigid links, the modules can reconfigure themselves spatially and position their sensors optimally as a function of the shape of the terrain, which may include walls with a negative slope. In the scope of the project, a final demonstration on a vertical cliff will validate the efficacy of the methods developed.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The MORPH project (FP 7, 2012-2016) aims at the development of efficient methods and tools to map the underwater environment in situations that defy existing technology. Namely, missions that involve underwater surveying and marine habitat mapping of rugged terrain and structures with full 3D complexity, including near-vertical cliffs. Potential applications include the study of cold water coral reefs or ecosystems in underwater canyons. For mapping purposes, the project advances the novel concept of an underwater robotic sensor carrier consisting of a number of spatially separated mobile robotic modules, each of them carrying complementary sensors. The modules are connected virtually via wireless communication links. Free from the constraints of rigid links, the modules can reconfigure themselves spatially and position their sensors optimally as a function of the shape of the terrain, which may include walls with a negative slope. In the scope of the project, a final demonstration on a vertical cliff will validate the efficacy of the methods developed.
2012
Belo, Felipe A. W.; Birk, Andreas; Brunskill, Christopher; Kirchner, Frank; Lappas, Vaios; Remy, C. David; Roccella, Stefano; Rossi, Claudio; Tikanmäki, Antti; Visentin, Gianfranco
The ESA Lunar Robotics Challenge: Simulating Operations at the Lunar South Pole Journal Article
In: Journal of Field Robotics, vol. 29, no. 4, pp. 601-626, 2012.
@article{ESA-LRC-JFR12,
title = {The ESA Lunar Robotics Challenge: Simulating Operations at the Lunar South Pole},
author = {Felipe A. W. Belo and Andreas Birk and Christopher Brunskill and Frank Kirchner and Vaios Lappas and C. David Remy and Stefano Roccella and Claudio Rossi and Antti Tikanmäki and Gianfranco Visentin},
url = {https://www.researchgate.net/publication/235633504_The_ESA_Lunar_Robotics_Challenge_Simulating_Operations_at_the_Lunar_South_Pole},
doi = {10.1002/rob.20429},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
journal = {Journal of Field Robotics},
volume = {29},
number = {4},
pages = {601-626},
abstract = {In 2008, the European Space Agency (ESA) challenged universities to design, develop, and test teleoperated robotic systems for a soil-sampling mission in a simulated lunar-crater-like environment. Eight teams participated and developed a wide range of engineering solutions that addressed the various technical and operational challenges posed by the unfavorable terrain and harsh environment. The robotic concepts developed by the teams are presented and evaluated in this paper. We highlight operational and technical issues that the teams experienced during an intensive 8-day field campaign, report on design solutions that were adopted to assist in operating a robotic system in a lunar environment, and describe the lesson learned through participation in this field-testing event.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In 2008, the European Space Agency (ESA) challenged universities to design, develop, and test teleoperated robotic systems for a soil-sampling mission in a simulated lunar-crater-like environment. Eight teams participated and developed a wide range of engineering solutions that addressed the various technical and operational challenges posed by the unfavorable terrain and harsh environment. The robotic concepts developed by the teams are presented and evaluated in this paper. We highlight operational and technical issues that the teams experienced during an intensive 8-day field campaign, report on design solutions that were adopted to assist in operating a robotic system in a lunar environment, and describe the lesson learned through participation in this field-testing event.
Birk, Andreas
A Corner for Educational Robotics Competitions Journal Article
In: IEEE Robotics & Automation Magazine, vol. 19, no. 1, pp. 21-21, 2012, ISSN: 1558-223X.
@article{RoboticsCompetitionsCorner-RAM12,
title = {A Corner for Educational Robotics Competitions},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/278232880_A_Corner_for_Educational_Robotics_Competitions},
doi = {10.1109/mra.2012.2186711},
issn = {1558-223X},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
journal = {IEEE Robotics & Automation Magazine},
volume = {19},
number = {1},
pages = {21-21},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Birk, Andreas; Pascoal, Antonio; Antonelli, Gianluca; Caiti, Andrea; Casalino, Giuseppe; Caffaz, Andrea
Cooperative Cognitive Control for Autonomous Underwater Vehicles (CO3AUVs): overview and progresses in the 3rd project year Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2012.
@inproceedings{Co3AUVs-NGCUV12,
title = {Cooperative Cognitive Control for Autonomous Underwater Vehicles (CO3AUVs): overview and progresses in the 3rd project year},
author = {Andreas Birk and Antonio Pascoal and Gianluca Antonelli and Andrea Caiti and Giuseppe Casalino and Andrea Caffaz},
url = {https://www.researchgate.net/publication/261990820_Cooperative_Cognitive_Control_for_Autonomous_Underwater_Vehicles_CO3AUVs_overview_and_progresses_in_the_3rd_project_year},
doi = {10.3182/20120410-3-PT-4028.00060},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {The paper describes the "Cooperative Cognitive Control for Autonomous Underwater Vehicles (Co3AUVs)" EU-project. This is a 7 th Framework Program STREP project under the theme: Information and Communication Technologies Cognitive Systems, Interaction, Robotics. The project duration is February 2009 to January 2012. The aim of Co3AUVs is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects are investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioral control strictly linked with the underwater communication challenges. This paper presents an overview of the project with a focus on results from the final 3rd project year.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper describes the "Cooperative Cognitive Control for Autonomous Underwater Vehicles (Co3AUVs)" EU-project. This is a 7 th Framework Program STREP project under the theme: Information and Communication Technologies Cognitive Systems, Interaction, Robotics. The project duration is February 2009 to January 2012. The aim of Co3AUVs is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects are investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioral control strictly linked with the underwater communication challenges. This paper presents an overview of the project with a focus on results from the final 3rd project year.
Birk, Andreas; Pfingsthorn, Max; Bülow, Heiko
Advances in Underwater Mapping and their Application Potential for Safety, Security, and Rescue Robotics (SSRR) Proceedings Article
In: IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR), IEEE Press, 2012.
@inproceedings{UnderwaterMappingOverview-SSRR12,
title = {Advances in Underwater Mapping and their Application Potential for Safety, Security, and Rescue Robotics (SSRR)},
author = {Andreas Birk and Max Pfingsthorn and Heiko Bülow},
url = {https://www.researchgate.net/publication/261229763_Advances_in_underwater_mapping_and_their_application_potential_for_Safety_Security_and_Rescue_Robotics_SSRR},
doi = {10.1109/SSRR.2012.6523907},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {We give here a short overview of recent contributions of the Jacobs Robotics group in underwater mapping. This work reaches up to 3D mapping of complex structures, which is of particular interest for Safety, Security, and Rescue Robotics (SSRR) missions with AUV, which tend to occur not in plain open waters but in complex environments. Examples of potential use cases include harbor security, infrastructure inspection after disasters, or military reconnaissance.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We give here a short overview of recent contributions of the Jacobs Robotics group in underwater mapping. This work reaches up to 3D mapping of complex structures, which is of particular interest for Safety, Security, and Rescue Robotics (SSRR) missions with AUV, which tend to occur not in plain open waters but in complex environments. Examples of potential use cases include harbor security, infrastructure inspection after disasters, or military reconnaissance.
Kalwa, Joerg; Pascoal, Antonio; Ridao, Pere; Birk, Andreas; Eichhorn, Mike; Brignone, Lorenzo
The European RnD-Project MORPH: Marine robotic systems of self-organizing, logically linked physical nodes Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2012.
@inproceedings{MORPH-NGCUV12,
title = {The European RnD-Project MORPH: Marine robotic systems of self-organizing, logically linked physical nodes},
author = {Joerg Kalwa and Antonio Pascoal and Pere Ridao and Andreas Birk and Mike Eichhorn and Lorenzo Brignone},
url = {https://www.researchgate.net/publication/265519733_The_European_RD-Project_MORPH_Marine_robotic_systems_of_self-organizing_logically_linked_physical_nodes},
doi = {10.3182/20120410-3-PT-4028.00058},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {The MORPH project (FP 7, 2012-2016) advances the novel concept of an underwater robotic system composed of a number of spatially separated mobile robot-modules, carrying complementary resources. Instead of being physically coupled, the modules are connected via communication links that rely on the flow of information among them, i. e. inter-module interactions are enabled by underwater communication networks at distant and close ranges and supported by visual perception at very close range. Without rigid links, the MSV can reconfigure itself and adapt in response to the shape of the terrain, including walls with negative slope. The MORPH concept requires qualitatively new behaviors such as adaptive sensor placement for perception and navigation, as well as environmental modeling in complex environments. On site view planning will lead to a solution well beyond the operational state of the art for underwater cliff surveys and other similar missions. A final demonstration on a vertical cliff, unfeasible automatically with today's technology, will validate the efficacy of the methods developed at the end of the project.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The MORPH project (FP 7, 2012-2016) advances the novel concept of an underwater robotic system composed of a number of spatially separated mobile robot-modules, carrying complementary resources. Instead of being physically coupled, the modules are connected via communication links that rely on the flow of information among them, i. e. inter-module interactions are enabled by underwater communication networks at distant and close ranges and supported by visual perception at very close range. Without rigid links, the MSV can reconfigure itself and adapt in response to the shape of the terrain, including walls with negative slope. The MORPH concept requires qualitatively new behaviors such as adaptive sensor placement for perception and navigation, as well as environmental modeling in complex environments. On site view planning will lead to a solution well beyond the operational state of the art for underwater cliff surveys and other similar missions. A final demonstration on a vertical cliff, unfeasible automatically with today's technology, will validate the efficacy of the methods developed at the end of the project.
Pathak, Kaustubh; Vaskevicius, Narunas; Bungiu, Francisc; Birk, Andreas
Utilizing Color Information in 3D Scan-Registration using Planar-Patches Matching Proceedings Article
In: IEEE International Conference on Multisensor Fusion and Information Integration (MFI), IEEE Press, 2012.
@inproceedings{ColorHistogramPlaneRegistration-MFI12,
title = {Utilizing Color Information in 3D Scan-Registration using Planar-Patches Matching},
author = {Kaustubh Pathak and Narunas Vaskevicius and Francisc Bungiu and Andreas Birk},
url = {https://www.researchgate.net/publication/261310811_Utilizing_color_information_in_3D_scan-registration_using_planar-patches_matching},
doi = {10.1109/MFI.2012.6343047},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IEEE International Conference on Multisensor Fusion and Information Integration (MFI)},
publisher = {IEEE Press},
abstract = {In previous work, the authors presented a 3D scan-registration algorithm based on minimizing the uncertainty-volume of the estimated inter-scan transform, computed by matching planar-patches extracted from a pair of 3D range-images. The method was shown to have a larger region of convergence than points-based methods like ICP. With the advent of newer sensors, color-information is now also available in addition to the depth-information in range-images. In this work, we show how this information can be exploited to make our algorithm computationally more efficient. The results are presented for two commercially available sensors providing color: the high-resolution, large field-of-view (FOV), slow scanning Faro sensor, and the low-resolution, small FOV, faster Kinect sensor.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In previous work, the authors presented a 3D scan-registration algorithm based on minimizing the uncertainty-volume of the estimated inter-scan transform, computed by matching planar-patches extracted from a pair of 3D range-images. The method was shown to have a larger region of convergence than points-based methods like ICP. With the advent of newer sensors, color-information is now also available in addition to the depth-information in range-images. In this work, we show how this information can be exploited to make our algorithm computationally more efficient. The results are presented for two commercially available sensors providing color: the high-resolution, large field-of-view (FOV), slow scanning Faro sensor, and the low-resolution, small FOV, faster Kinect sensor.
Pfingsthorn, Max; Birk, Andreas; Bülow, Heiko
Uncertainty Estimation for a 6-DoF Spectral Registration method as basis for Sonar-based Underwater 3D SLAM Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2012.
@inproceedings{SpectralUnderwaterSLAM-ICRA12,
title = {Uncertainty Estimation for a 6-DoF Spectral Registration method as basis for Sonar-based Underwater 3D SLAM},
author = {Max Pfingsthorn and Andreas Birk and Heiko Bülow},
url = {https://www.researchgate.net/publication/239763466_Uncertainty_estimation_for_a_6DoF_spectral_registration_method_as_basis_for_sonar-based_underwater_3D_SLAM},
doi = {10.1109/ICRA.2012.6224731},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {An uncertainty estimation method for 6 degree of freedom (6-DoF) spectral registration is introduced here. The underlying 6-DoF registration method based on Phase Only Matched Filtering (POMF) is capable of dealing with very noisy sensor data. It is hence well suited for 3D underwater mapping, where relatively inaccurate sonar imaging devices have to be employed. An uncertainty estimation method is required to use this registration method in a Simultaneous Localization and Mapping (SLAM) framework. To our knowledge, the first such method for 6-DoF spectral registration is presented here. This new uncertainty estimation method treats the POMF results as probability mass functions (PMF). Due to the decoupling in the underlying method, yaw is computed by a one-dimensional POMF leading hence to a 1D PMF; roll and pitch are simultaneously computed and hence encoded in a 2D PMF. Furthermore, a 3D PMF is generated for the translation as it is determined by a 3D POMF. A normal distribution is fitted on each of the PMF to get the uncertainty estimate. The method is experimentally evaluated with simulated as well as real world sonar data. It is shown that it indeed can be used for SLAM, which significantly improves the map quality.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An uncertainty estimation method for 6 degree of freedom (6-DoF) spectral registration is introduced here. The underlying 6-DoF registration method based on Phase Only Matched Filtering (POMF) is capable of dealing with very noisy sensor data. It is hence well suited for 3D underwater mapping, where relatively inaccurate sonar imaging devices have to be employed. An uncertainty estimation method is required to use this registration method in a Simultaneous Localization and Mapping (SLAM) framework. To our knowledge, the first such method for 6-DoF spectral registration is presented here. This new uncertainty estimation method treats the POMF results as probability mass functions (PMF). Due to the decoupling in the underlying method, yaw is computed by a one-dimensional POMF leading hence to a 1D PMF; roll and pitch are simultaneously computed and hence encoded in a 2D PMF. Furthermore, a 3D PMF is generated for the translation as it is determined by a 3D POMF. A normal distribution is fitted on each of the PMF to get the uncertainty estimate. The method is experimentally evaluated with simulated as well as real world sonar data. It is shown that it indeed can be used for SLAM, which significantly improves the map quality.
Rathnam, Ravi; Birk, Andreas
Cooperative 3D Exploration under Communication Constraints Proceedings Article
In: IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV), 2012.
@inproceedings{UnderwaterCoopExploration-NGCUV12,
title = {Cooperative 3D Exploration under Communication Constraints},
author = {Ravi Rathnam and Andreas Birk},
url = {https://www.researchgate.net/publication/260292078_Cooperative_3D_Exploration_under_Communication_Constraints},
doi = {10.3182/20120410-3-PT-4028.00016},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IFAC Workshop on Navigation, Guidance and Control of Underwater Vehicles (NGCUV)},
abstract = {In this paper, a 3d exploration algorithm is presented which always keeps the robots within communication range of each other. The exploration is integrated into a control infrastructure containing sensors, mapping, navigation, and control of actuators. The algorithm is tested in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. Furthermore, we test the effect of communication range, and sensor range on the exploration.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, a 3d exploration algorithm is presented which always keeps the robots within communication range of each other. The exploration is integrated into a control infrastructure containing sensors, mapping, navigation, and control of actuators. The algorithm is tested in a high fidelity simulator which takes into account the dynamics of the robot, and simulates the required sensors. Furthermore, we test the effect of communication range, and sensor range on the exploration.
Schwertfeger, Sören; Birk, Andreas
A Short Overview of Recent Advances in Map Evaluation Proceedings Article
In: IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR), IEEE Press, 2012.
@inproceedings{MapEvaluationOverview-SSRR12,
title = {A Short Overview of Recent Advances in Map Evaluation},
author = {Sören Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/261163466_A_short_overview_of_recent_advances_in_map_evaluation},
doi = {10.1109/SSRR.2012.6523906},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {IEEE International Symposium on Safety, Security, Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {Mapping is an important task for mobile robots in general and in Safety, Security, and Rescue Robotics (SSRR) in particular - often maps are even a core mission deliverable for SSRR applications. The assessment of the quality of maps in a simple, efficient and automated way is hence of high interest. But it is not trivial and an ongoing research topic. Here, an overview of advances with a new approach on map evaluation is presented. This structure-based method makes use of a Topology Graph, a topological, abstracted representation of the map. The Topology Graph is constructed by generating and processing a Voronoi Diagram from a 2D grid map. Having a ground truth map, the Topology Graphs of both maps are matched using both similarity metrics on the vertices as well as structural matching of subgraph isomorphisms.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mapping is an important task for mobile robots in general and in Safety, Security, and Rescue Robotics (SSRR) in particular - often maps are even a core mission deliverable for SSRR applications. The assessment of the quality of maps in a simple, efficient and automated way is hence of high interest. But it is not trivial and an ongoing research topic. Here, an overview of advances with a new approach on map evaluation is presented. This structure-based method makes use of a Topology Graph, a topological, abstracted representation of the map. The Topology Graph is constructed by generating and processing a Voronoi Diagram from a 2D grid map. Having a ground truth map, the Topology Graphs of both maps are matched using both similarity metrics on the vertices as well as structural matching of subgraph isomorphisms.
Vaskevicius, Narunas; Pathak, Kaustubh; Ichim, Alexandru-Eugen; Birk, Andreas
The Jacobs Robotics Approach to Object Recognition and Localization in the Context of the ICRA'11 Solutions in Perception Challenge Proceedings Article
In: International Conference on Robotics and Automation (ICRA), IEEE Press, 2012.
@inproceedings{PerceptionChallenge11-JacobsTeam-ICRA12,
title = {The Jacobs Robotics Approach to Object Recognition and Localization in the Context of the ICRA'11 Solutions in Perception Challenge},
author = {Narunas Vaskevicius and Kaustubh Pathak and Alexandru-Eugen Ichim and Andreas Birk},
url = {https://www.researchgate.net/publication/254041356_The_Jacobs_Robotics_approach_to_object_recognition_and_localization_in_the_context_of_the_ICRA'11_Solutions_in_Perception_Challenge},
doi = {10.1109/ICRA.2012.6225335},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {In this paper, we give an overview of the Jacobs Robotics entry to the ICRA'11 Solutions in Perception Challenge. We present our multi-pronged strategy for object recognition and localization based on the integrated geometric and visual information available from the Kinect sensor. Firstly, the range image is over-segmented using an edge-detection algorithm and regions of interest are extracted based on a simple shape-analysis per segment. Then, these selected regions of the scene are matched with known objects using visual features and their distribution in 3D space. Finally, generated hypotheses about the positions of the objects are tested by back-projecting learned 3D models to the scene using estimated transformations and sensor model. Our method won the second place among eight competing algorithms, only marginally losing to the winner.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, we give an overview of the Jacobs Robotics entry to the ICRA'11 Solutions in Perception Challenge. We present our multi-pronged strategy for object recognition and localization based on the integrated geometric and visual information available from the Kinect sensor. Firstly, the range image is over-segmented using an edge-detection algorithm and regions of interest are extracted based on a simple shape-analysis per segment. Then, these selected regions of the scene are matched with known objects using visual features and their distribution in 3D space. Finally, generated hypotheses about the positions of the objects are tested by back-projecting learned 3D models to the scene using estimated transformations and sensor model. Our method won the second place among eight competing algorithms, only marginally losing to the winner.
2011
Bülow, Heiko; Birk, Andreas
Spectral Registration of Noisy Sonar Data for Underwater 3D Mapping Journal Article
In: Autonomous Robots, vol. 30, no. 3, pp. 307-331, 2011.
@article{Spectral3DSonarMap-AuRo11,
title = {Spectral Registration of Noisy Sonar Data for Underwater 3D Mapping},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/220474258_Spectral_registration_of_noisy_sonar_data_for_underwater_3D_Mapping},
doi = {10.1007/s10514-011-9221-8},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
journal = {Autonomous Robots},
volume = {30},
number = {3},
pages = {307-331},
abstract = {3D mapping is very challenging in the underwater domain, especially due to the lack of high resolution, low noise sensors. A new spectral registration method is presented that can determine the spatial 6 DOF transformation between pairs of very noisy 3D scans with only partial overlap. The approach is hence suited to cope with sonar as the predominant underwater sensor. The spectral registration method is based on Phase Only Matched Filtering (POMF) on non-trivially resampled spectra of the 3D data. Two extensive sets of experiments are presented. First, evaluations with simulated data are done where the type and amount of noise can be controlled and the ground truth transformations between scans are known. Second, real world data from a Tritech Eclipse sonar is used. Concretely, 18 sonar scans of a large structure in form of a flood gate and a lock in the river Lesum in Bremen are used for 3D mapping. In doing so, the spectral registration method is compared to two other methods suited for noisy 3D registrations, namely Iterative Closest Point (ICP) and plane-based registration. It is shown that the spectral registration method performs very well in terms of the resulting 3D map as well as its run-times.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
3D mapping is very challenging in the underwater domain, especially due to the lack of high resolution, low noise sensors. A new spectral registration method is presented that can determine the spatial 6 DOF transformation between pairs of very noisy 3D scans with only partial overlap. The approach is hence suited to cope with sonar as the predominant underwater sensor. The spectral registration method is based on Phase Only Matched Filtering (POMF) on non-trivially resampled spectra of the 3D data. Two extensive sets of experiments are presented. First, evaluations with simulated data are done where the type and amount of noise can be controlled and the ground truth transformations between scans are known. Second, real world data from a Tritech Eclipse sonar is used. Concretely, 18 sonar scans of a large structure in form of a flood gate and a lock in the river Lesum in Bremen are used for 3D mapping. In doing so, the spectral registration method is compared to two other methods suited for noisy 3D registrations, namely Iterative Closest Point (ICP) and plane-based registration. It is shown that the spectral registration method performs very well in terms of the resulting 3D map as well as its run-times.
Bülow, Heiko; Birk, Andreas
Spectral Registration of Volume Data for 6-DOF Spatial Transformations plus Scale Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE press, 2011.
@inproceedings{7-DoF-registration-ICRA11,
title = {Spectral Registration of Volume Data for 6-DOF Spatial Transformations plus Scale},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/224252494_Spectral_registration_of_volume_data_for_6-DOF_spatial_transformations_plus_scale},
doi = {10.1109/ICRA.2011.5980009},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE press},
abstract = {This paper presents a robust spectral based 6 Degrees Of Freedoms (DOF) registration method for 3D data which is in addition capable to determine the scaling of the registered object. The principle is based on decoupling 3D rotation from 3D translation by a corresponding resampling process of the spectral magnitude using a 3D Fast Fourier Transform (FFT), a SO(3) Fourier transform and an integrated Mellin transform for a scale determination. The registration is then subsequently carried out, first the registration of rotation is done, in a second step the scale is determined and finally the translation is calculated. Results are presented with generated test series on public available data and data recorded with Laser Range Finders (LRF). Furthermore one example of a scan pair recorded under realistic conditions showing the robustness of this method is given.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a robust spectral based 6 Degrees Of Freedoms (DOF) registration method for 3D data which is in addition capable to determine the scaling of the registered object. The principle is based on decoupling 3D rotation from 3D translation by a corresponding resampling process of the spectral magnitude using a 3D Fast Fourier Transform (FFT), a SO(3) Fourier transform and an integrated Mellin transform for a scale determination. The registration is then subsequently carried out, first the registration of rotation is done, in a second step the scale is determined and finally the translation is calculated. Results are presented with generated test series on public available data and data recorded with Laser Range Finders (LRF). Furthermore one example of a scan pair recorded under realistic conditions showing the robustness of this method is given.
Birk, Andreas
What is Robotics? An interdisciplinary field is getting even more diverse Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 18, no. 4, pp. 94-95, 2011.
@article{EduColumn-RoboticsGettingMoreDiverse-RAM11,
title = {What is Robotics? An interdisciplinary field is getting even more diverse},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/220556132_What_Is_Robotics_An_Interdisciplinary_Field_Is_Getting_Even_More_Diverse},
doi = {10.1109/MRA.2011.943235},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {18},
number = {4},
pages = {94-95},
abstract = {Robotics is a very young field, which is quite interdisciplinary due to its very nature of developing and building systems that – following its science fiction roots that preceded the real technology – combine electrical and mechanical “body” components with computer “brains”. Accordingly, mechanical engineering, electrical engineering, and computer science are core disciplines involved in robotics. This already makes robotics education quite challenging. There are some natural differences in the curricula of these three disciplines, especially when it comes to the typical coverage of mathematical topics during undergraduate and graduate studies. But fortunately, the borders between these different disciplines have become much more dispersed in recent years. Robots have become popular tools in CS education at many places; not necessarily to just teach robotics but for the more general goal to educate CS students about a variety of physical system aspects. At the same time, mechanical and electrical engineering programs nowadays include substantial amounts of CS related courses. I even dare to claim that not only robotics but all engineering fields have nowadays become strongly software-driven and hence CS aspects are a significant part of their education activities. But we are now facing a new, more radical paradigm shift, namely an increasing relevance of bio/cognitive disciplines in robotics.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Robotics is a very young field, which is quite interdisciplinary due to its very nature of developing and building systems that – following its science fiction roots that preceded the real technology – combine electrical and mechanical “body” components with computer “brains”. Accordingly, mechanical engineering, electrical engineering, and computer science are core disciplines involved in robotics. This already makes robotics education quite challenging. There are some natural differences in the curricula of these three disciplines, especially when it comes to the typical coverage of mathematical topics during undergraduate and graduate studies. But fortunately, the borders between these different disciplines have become much more dispersed in recent years. Robots have become popular tools in CS education at many places; not necessarily to just teach robotics but for the more general goal to educate CS students about a variety of physical system aspects. At the same time, mechanical and electrical engineering programs nowadays include substantial amounts of CS related courses. I even dare to claim that not only robotics but all engineering fields have nowadays become strongly software-driven and hence CS aspects are a significant part of their education activities. But we are now facing a new, more radical paradigm shift, namely an increasing relevance of bio/cognitive disciplines in robotics.
Birk, Andreas; Antonelli, Gianluca; Caiti, Andrea; Casalino, Giuseppe; Indiveri, Giovanni; Pascoal, Antonio; Caffaz, Andrea
The Co3-AUVs (COoperative COgnitive COntrol for Autonomous Underwater Vehicles) Project: overview and current progresses Proceedings Article
In: IEEE Oceans, 2011.
@inproceedings{Co3AUVs-EuroOceans11,
title = {The Co3-AUVs (COoperative COgnitive COntrol for Autonomous Underwater Vehicles) Project: overview and current progresses},
author = {Andreas Birk and Gianluca Antonelli and Andrea Caiti and Giuseppe Casalino and Giovanni Indiveri and Antonio Pascoal and Andrea Caffaz},
url = {https://www.researchgate.net/publication/224255692_The_CO3AUVs_Cooperative_Cognitive_Control_for_Autonomous_Underwater_Vehicles_project_Overview_and_current_progresses},
doi = {10.1109/Oceans-Spain.2011.6003552},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE Oceans},
abstract = {“Cooperative Cognitive Control for Autonomous Underwater Vehicles” (Co3AUVs) is Collaborative Project (STREP) funded by the European Commission (EC) under the Seventh Framework Programme (FP7), “Information and Communication technologies” (ICT) Challenge 2: “Cognitive Systems, Interaction, Robotics”. The aim of Co3AUVs is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects are investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioral control strictly linked with the underwater communication challenges. This paper presents results from the first two years of the project, including the simulator, the development of vehicles and components, 2D as well as 3D underwater mapping, cooperative navigation and motion control, and cooperative skills.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
“Cooperative Cognitive Control for Autonomous Underwater Vehicles” (Co3AUVs) is Collaborative Project (STREP) funded by the European Commission (EC) under the Seventh Framework Programme (FP7), “Information and Communication technologies” (ICT) Challenge 2: “Cognitive Systems, Interaction, Robotics”. The aim of Co3AUVs is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects are investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioral control strictly linked with the underwater communication challenges. This paper presents results from the first two years of the project, including the simulator, the development of vehicles and components, 2D as well as 3D underwater mapping, cooperative navigation and motion control, and cooperative skills.
Birk, Andreas; Wiggerich, Burkhard; Bülow, Heiko; Pfingsthorn, Max; Schwertfeger, Soeren
In: Journal of Intelligent and Robotic Systems, vol. 64, no. 1, pp. 57-76, 2011.
@article{JIRS-UAVmosaic-ELROB-RREE-2011,
title = {Safety, Security, and Rescue Missions with an Unmanned Aerial Vehicle (UAV): Aerial Mosaicking and Autonomous Flight at the 2009 European Land Robots Trials (ELROB) and the 2010 Response Robot Evaluation Exercises (RREE)},
author = {Andreas Birk and Burkhard Wiggerich and Heiko Bülow and Max Pfingsthorn and Soeren Schwertfeger},
url = {https://www.researchgate.net/publication/227213785_Safety_Security_and_Rescue_Missions_with_an_Unmanned_Aerial_Vehicle_UAV},
doi = {10.1007/s10846-011-9546-8},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
journal = {Journal of Intelligent and Robotic Systems},
volume = {64},
number = {1},
pages = {57-76},
abstract = {Several missions with an Unmanned Aerial Vehicle (UAV) in different realistic safety, security, and rescue field tests are presented. First, results from two safety and security missions at the 2009 European Land Robot Trials (ELROB) are presented. A UAV in form of an Airrobot AR100-B is used in a reconnaissance and in a camp security scenario. The UAV is capable of autonomous waypoint navigation using onboard GPS processing. A digital video stream from the vehicle is used to create photo maps—also known as mosaicking—in real time at the operator station. This mapping is done using an enhanced version of Fourier Mellin based registration, which turns out to be very fast and robust. Furthermore, results from a rescue oriented scenario at the 2010 Response Robot Evaluation Exercises (RREE) at Disaster City, Texas are presented. The registration for the aerial mosaicking is supplemented by an uncertainty metric and embedded into Simultaneous Localization and Mapping (SLAM), which further enhances the photo maps as main mission deliveries.
Keywords: Unmanned Aerial Vehicle (UAV), Micro Aerial Vehicle (MAV), Drone, Autonomy, Registration, Mosaicking, Visual SLAM, Exploration, Reconnaissance},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Several missions with an Unmanned Aerial Vehicle (UAV) in different realistic safety, security, and rescue field tests are presented. First, results from two safety and security missions at the 2009 European Land Robot Trials (ELROB) are presented. A UAV in form of an Airrobot AR100-B is used in a reconnaissance and in a camp security scenario. The UAV is capable of autonomous waypoint navigation using onboard GPS processing. A digital video stream from the vehicle is used to create photo maps—also known as mosaicking—in real time at the operator station. This mapping is done using an enhanced version of Fourier Mellin based registration, which turns out to be very fast and robust. Furthermore, results from a rescue oriented scenario at the 2010 Response Robot Evaluation Exercises (RREE) at Disaster City, Texas are presented. The registration for the aerial mosaicking is supplemented by an uncertainty metric and embedded into Simultaneous Localization and Mapping (SLAM), which further enhances the photo maps as main mission deliveries.
Keywords: Unmanned Aerial Vehicle (UAV), Micro Aerial Vehicle (MAV), Drone, Autonomy, Registration, Mosaicking, Visual SLAM, Exploration, Reconnaissance
Keywords: Unmanned Aerial Vehicle (UAV), Micro Aerial Vehicle (MAV), Drone, Autonomy, Registration, Mosaicking, Visual SLAM, Exploration, Reconnaissance
Buelow, Heiko; Birk, Andreas
Diver Detection by Motion-Segmentation and Shape-Analysis from a Moving Vehicle Proceedings Article
In: IEEE Oceans, 2011.
@inproceedings{DiverDetection-Oceans11,
title = {Diver Detection by Motion-Segmentation and Shape-Analysis from a Moving Vehicle},
author = {Heiko Buelow and Andreas Birk},
url = {https://www.researchgate.net/publication/254013127_Diver_detection_by_motion-segmentation_and_shape-analysis_from_a_moving_vehicle},
doi = {10.23919/OCEANS.2011.6107115},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE Oceans},
abstract = {The operation of Autonomous Underwater Vehicles (AUV) in scenarios where also human divers are working, e.g., during archaeological missions or during inspection and service tasks, bears an interesting application potential. But according scenarios also require the detection of the divers for at least safety reasons. Diver detection is also of interest for other applications, e.g., within the context of harbor security scenarios. In this paper, an approach for diver detection with a monocular camera on a moving vehicle is presented. The main challenge is the motion compensation in the video stream, which is achieved here by a spectral registration method in form of the improved Fourier Mellin Invariant (iFMI) registration.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The operation of Autonomous Underwater Vehicles (AUV) in scenarios where also human divers are working, e.g., during archaeological missions or during inspection and service tasks, bears an interesting application potential. But according scenarios also require the detection of the divers for at least safety reasons. Diver detection is also of interest for other applications, e.g., within the context of harbor security scenarios. In this paper, an approach for diver detection with a monocular camera on a moving vehicle is presented. The main challenge is the motion compensation in the video stream, which is achieved here by a spectral registration method in form of the improved Fourier Mellin Invariant (iFMI) registration.
Buelow, Heiko; Birk, Andreas
Gesture-Recognition as basis for a Human Robot Interface (HRI) on a AUV Proceedings Article
In: IEEE Oceans, 2011.
@inproceedings{GestureRecognition-Oceans11,
title = {Gesture-Recognition as basis for a Human Robot Interface (HRI) on a AUV},
author = {Heiko Buelow and Andreas Birk},
url = {https://www.researchgate.net/publication/254013131_Gesture-recognition_as_basis_for_a_Human_Robot_Interface_HRI_on_a_AUV},
doi = {10.23919/OCEANS.2011.6107118},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE Oceans},
abstract = {An approach for gesture detection with a monocular camera on a moving vehicle is presented in this paper. The work consists of two parts. The first part deals with the video image processing. The main challenge here is the compensation of the ego-motion of the vehicle by image registration, which is required to facilitate the subsequent segmentation of relevant motions of the gestures. This is done with a spectral registration method, which proves to be useful for underwater scenarios. The second part deals with the recognition of gestures using motion trajectories. State machines where the different states are associated with the directions of movements are used for this purpose.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
An approach for gesture detection with a monocular camera on a moving vehicle is presented in this paper. The work consists of two parts. The first part deals with the video image processing. The main challenge here is the compensation of the ego-motion of the vehicle by image registration, which is required to facilitate the subsequent segmentation of relevant motions of the gestures. This is done with a spectral registration method, which proves to be useful for underwater scenarios. The second part deals with the recognition of gestures using motion trajectories. State machines where the different states are associated with the directions of movements are used for this purpose.
Pfingsthorn, Max; Birk, Andreas; Vaskevicius, Narunas
Semantic Annotation of Ground and Vegetation Types in 3D Maps for Autonomous Underwater Vehicle Operation Proceedings Article
In: IEEE Oceans, 2011.
@inproceedings{SemanticUnderwater3DMapping-Oceans11,
title = {Semantic Annotation of Ground and Vegetation Types in 3D Maps for Autonomous Underwater Vehicle Operation},
author = {Max Pfingsthorn and Andreas Birk and Narunas Vaskevicius},
url = {https://www.researchgate.net/publication/254013137_Semantic_annotation_of_ground_and_vegetation_types_in_3D_maps_for_autonomous_underwater_vehicle_operation},
doi = {10.23919/OCEANS.2011.6107122},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE Oceans},
abstract = {The semantic annotation of 3D maps generated by an Autonomous Underwater Vehicle (AUV) is presented. Two different methods are used for this purpose. First, a fitting of large planar patches plus an analysis of the plane normals is proposed. Second, a local analysis of the normals on the point cloud level is employed. Both methods complement each other. While the first captures large scale environment structures like the sea floor, cliffs, and (man-made) walls, the second is targeted at smaller, locally non-planar, elements like vegetation and rocks. The semantic 3D mapping is evaluated in a high-fidelity simulator where it is shown that the two methods are very fast and work as intended.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The semantic annotation of 3D maps generated by an Autonomous Underwater Vehicle (AUV) is presented. Two different methods are used for this purpose. First, a fitting of large planar patches plus an analysis of the plane normals is proposed. Second, a local analysis of the normals on the point cloud level is employed. Both methods complement each other. While the first captures large scale environment structures like the sea floor, cliffs, and (man-made) walls, the second is targeted at smaller, locally non-planar, elements like vegetation and rocks. The semantic 3D mapping is evaluated in a high-fidelity simulator where it is shown that the two methods are very fast and work as intended.
Pfingsthorn, Max; Birk, Andreas; Vaskevicius, Narunas; Pathak, Kaustubh
Cooperative 3D Mapping under Underwater Communication Constraints Proceedings Article
In: IEEE Oceans, 2011.
@inproceedings{CooperativeUnderwater3DMapping-Oceans11,
title = {Cooperative 3D Mapping under Underwater Communication Constraints},
author = {Max Pfingsthorn and Andreas Birk and Narunas Vaskevicius and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/254013139_Cooperative_3D_mapping_under_underwater_communication_constraints},
doi = {10.23919/OCEANS.2011.6107123},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE Oceans},
abstract = {Marine robots are so far mainly restricted to usage in open sea application where 2D maps are sufficient, however, full 3D information becomes more and more important in complex scenarios. These include complex inspection, construction, or maintenance missions that involve non-trivial structures as for example in oil- and gas-production and transport, underwater archeology, or harbor applications. In this paper, we present a graph based multi-robot mapping method that requires minimal communication bandwidth between the robots. We show that using this method, it is feasible to construct a large scale map under realistic bandwidth constraints imposed by available underwater sonic modems.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Marine robots are so far mainly restricted to usage in open sea application where 2D maps are sufficient, however, full 3D information becomes more and more important in complex scenarios. These include complex inspection, construction, or maintenance missions that involve non-trivial structures as for example in oil- and gas-production and transport, underwater archeology, or harbor applications. In this paper, we present a graph based multi-robot mapping method that requires minimal communication bandwidth between the robots. We show that using this method, it is feasible to construct a large scale map under realistic bandwidth constraints imposed by available underwater sonic modems.
Poppinga, Jann; Birk, Andreas; Pathak, Kaustubh; Vaskevicius, Narunas
Fast 6-DOF Path Planning for Autonomous Underwater Vehicles (AUV) based on 3D Plane Mapping Proceedings Article
In: IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pp. 1-6, IEEE Press, 2011.
@inproceedings{SSRR11-AUV-3D-pathplan,
title = {Fast 6-DOF Path Planning for Autonomous Underwater Vehicles (AUV) based on 3D Plane Mapping},
author = {Jann Poppinga and Andreas Birk and Kaustubh Pathak and Narunas Vaskevicius},
url = {https://www.researchgate.net/publication/254013483_Fast_6DOF_path_planning_for_Autonomous_Underwater_Vehicles_AUV_based_on_3D_plane_mapping},
doi = {10.1109/SSRR.2011.6106771},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)},
pages = {1-6},
publisher = {IEEE Press},
abstract = {A method for 6 degree of freedom (6-DOF) path-planning for Autonomous Underwater Vehicles (AUV) is presented. It is based on an augmentation of Rapidly-exploring Random Trees (RRT) and the Probabilistic Roadmap Method (PRM) with a plane-fitting method as part of 3D underwater mapping. This is of particular interest for Safety, Security, and Rescue Robotics (SSRR) missions with AUV, which tend to occur not in plain open waters but in complex environments. Examples include harbor security, infrastructure inspection after disasters, and military reconnaissance. The special representation of the 3D map used here allows for significantly faster generation of paths than with state of the art point clouds. The method is evaluated with experiments using real world data from a Tritech Eclipse sonar. We present results of 6-DOF path-planning through a 3D map that is generated in a larger underwater structure, namely the Lesumer Sperrwerk, a flood gate in the river Lesum in Bremen, Germany.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A method for 6 degree of freedom (6-DOF) path-planning for Autonomous Underwater Vehicles (AUV) is presented. It is based on an augmentation of Rapidly-exploring Random Trees (RRT) and the Probabilistic Roadmap Method (PRM) with a plane-fitting method as part of 3D underwater mapping. This is of particular interest for Safety, Security, and Rescue Robotics (SSRR) missions with AUV, which tend to occur not in plain open waters but in complex environments. Examples include harbor security, infrastructure inspection after disasters, and military reconnaissance. The special representation of the 3D map used here allows for significantly faster generation of paths than with state of the art point clouds. The method is evaluated with experiments using real world data from a Tritech Eclipse sonar. We present results of 6-DOF path-planning through a 3D map that is generated in a larger underwater structure, namely the Lesumer Sperrwerk, a flood gate in the river Lesum in Bremen, Germany.
Rathnam, Ravi; Birk, Andreas
Distributed Communicative Exploration under Underwater Communication Constraints Proceedings Article
In: IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pp. 1-6, IEEE Press, 2011.
@inproceedings{SSRR11-DistributedCommExplore,
title = {Distributed Communicative Exploration under Underwater Communication Constraints},
author = {Ravi Rathnam and Andreas Birk},
url = {https://www.researchgate.net/publication/241625353_Distributed_Communicative_Exploration_under_underwater_communication_constraints},
doi = {10.1109/SSRR.2011.6106767},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)},
pages = {1-6},
publisher = {IEEE Press},
abstract = {Exploration by Unmanned Underwater Vehicles (UUV) is of interest in different application areas of Safety, Security, and Rescue Robotics (SSRR) including infrastructure inspection after disasters, military reconnaissance, marine search and rescue, and mine hunting. A multi-robot system performing exploration can be highly beneficial for this purpose by distributing the work amongst the different robots. In this paper, we present a distributed approach to multi-robot exploration which always keeps all the robots in communication range. The algorithm is based on a utility function that guarantees communication while using best effort for the exploration itself. The robots randomly sample configuration changes of the overall system, i.e., the set of the next best joined movements based on this utility function. This process can be very well distributed with very few communication overhead by having each robot computing a part of the sample and broadcasting its best candidate configuration change found. The globally best candidate among the broadcasted ones is then executed by the team. The algorithm is tested in a high-fidelity marine robotics simulator including realistic vehicle physics and proper underlying software for mapping, path-planning, and motion-control.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Exploration by Unmanned Underwater Vehicles (UUV) is of interest in different application areas of Safety, Security, and Rescue Robotics (SSRR) including infrastructure inspection after disasters, military reconnaissance, marine search and rescue, and mine hunting. A multi-robot system performing exploration can be highly beneficial for this purpose by distributing the work amongst the different robots. In this paper, we present a distributed approach to multi-robot exploration which always keeps all the robots in communication range. The algorithm is based on a utility function that guarantees communication while using best effort for the exploration itself. The robots randomly sample configuration changes of the overall system, i.e., the set of the next best joined movements based on this utility function. This process can be very well distributed with very few communication overhead by having each robot computing a part of the sample and broadcasting its best candidate configuration change found. The globally best candidate among the broadcasted ones is then executed by the team. The algorithm is tested in a high-fidelity marine robotics simulator including realistic vehicle physics and proper underlying software for mapping, path-planning, and motion-control.
Schwertfeger, Soeren; Birk, Andreas; Bülow, Heiko
Using iFMI Spectral Registration for Video Stabilization and Motion Detection by an Unmanned Aerial Vehicle (UAV) Proceedings Article
In: IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pp. 1-6, IEEE Press, 2011.
@inproceedings{SSRR11-UAV-MotionDetection,
title = {Using iFMI Spectral Registration for Video Stabilization and Motion Detection by an Unmanned Aerial Vehicle (UAV)},
author = {Soeren Schwertfeger and Andreas Birk and Heiko Bülow},
url = {https://www.researchgate.net/publication/254013482_Using_iFMI_spectral_registration_for_video_stabilization_and_motion_detection_by_an_Unmanned_Aerial_Vehicle_UAV},
doi = {10.1109/SSRR.2011.6106770},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)},
pages = {1-6},
publisher = {IEEE Press},
abstract = {Unmanned Aerial Vehicles (UAV), especially in the form of Micro Aerial Vehicles (MAV) are useful tools for reconnaissance, surveillance, and general situation assessment in safety, security, and rescue missions. Many UAV have meanwhile good autonomous flight capabilities, especially by tracking pre-planned routes via GPS or for station-keeping. Here it is shown how the video stream from an UAV can be analyzed to automatically detect motion in the scene while the vehicle is moving itself. Concretely, it is shown how a spectral image registration method, the improved Fourier Mellin Invariant method (iFMI), can be used for video stabilization and motion segmentation. The method is first analyzed with scenes containing artificial markers for ground truth evaluation. Furthermore, results from aerial video data from a quadcopter platform are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Unmanned Aerial Vehicles (UAV), especially in the form of Micro Aerial Vehicles (MAV) are useful tools for reconnaissance, surveillance, and general situation assessment in safety, security, and rescue missions. Many UAV have meanwhile good autonomous flight capabilities, especially by tracking pre-planned routes via GPS or for station-keeping. Here it is shown how the video stream from an UAV can be analyzed to automatically detect motion in the scene while the vehicle is moving itself. Concretely, it is shown how a spectral image registration method, the improved Fourier Mellin Invariant method (iFMI), can be used for video stabilization and motion segmentation. The method is first analyzed with scenes containing artificial markers for ground truth evaluation. Furthermore, results from aerial video data from a quadcopter platform are presented.
Schwertfeger, Soeren; Jacoff, Adam; Pellenz, Johannes; Birk, Andreas
Using a Fiducial Map Metric for Assessing Map Quality in the context of RoboCup Rescue Proceedings Article
In: IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), pp. 1-6, IEEE Press, 2011.
@inproceedings{SSRR11-FiducialMapMetric,
title = {Using a Fiducial Map Metric for Assessing Map Quality in the context of RoboCup Rescue},
author = {Soeren Schwertfeger and Adam Jacoff and Johannes Pellenz and Andreas Birk},
url = {https://www.researchgate.net/publication/254013822_Using_a_fiducial_map_metric_for_assessing_map_quality_in_the_context_of_RoboCup_Rescue},
doi = {10.1109/SSRR.2011.6106762},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
booktitle = {IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)},
pages = {1-6},
publisher = {IEEE Press},
abstract = {Mapping is an important task for mobile robots in general and for Safety, Security, and Rescue Robotics (SSRR) in particular. It is hence one core aspect which is evaluated in the RoboCup Rescue league. But assessing the quality of maps in a simple and efficient way is not trivial, especially if no detailed, complete ground truth data of the environment is available. A new approach on map evaluation is presented here. It makes use of artificial objects placed in the environment named “fiducials”. Using the known ground-truth positions and the positions of the fiducials identified in the map, a number of quality attributes can be assigned to that map. Depending on the application domain those attributes can weighed to compute a final score. Results are presented that are based on using this method during the RoboCup Rescue competition 2010 in Singapore where maps were generated by different teams in an maze populated with fiducials. Those maps are evaluated here and compared to a human judgment, showing the effectiveness of the fiducial approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mapping is an important task for mobile robots in general and for Safety, Security, and Rescue Robotics (SSRR) in particular. It is hence one core aspect which is evaluated in the RoboCup Rescue league. But assessing the quality of maps in a simple and efficient way is not trivial, especially if no detailed, complete ground truth data of the environment is available. A new approach on map evaluation is presented here. It makes use of artificial objects placed in the environment named “fiducials”. Using the known ground-truth positions and the positions of the fiducials identified in the map, a number of quality attributes can be assigned to that map. Depending on the application domain those attributes can weighed to compute a final score. Results are presented that are based on using this method during the RoboCup Rescue competition 2010 in Singapore where maps were generated by different teams in an maze populated with fiducials. Those maps are evaluated here and compared to a human judgment, showing the effectiveness of the fiducial approach.
Vaskevicius, Narunas; Birk, Andreas
Towards Pathplanning for Unmanned Ground Vehicles (UGV) in 3D Plane-Maps of Unstructured Environments Journal Article
In: KI - Kuenstliche Intelligenz, vol. 25, no. 2, pp. 141-144, 2011.
@article{3D-UGVpathplan-KI11,
title = {Towards Pathplanning for Unmanned Ground Vehicles (UGV) in 3D Plane-Maps of Unstructured Environments},
author = {Narunas Vaskevicius and Andreas Birk},
url = {https://www.researchgate.net/publication/220633391_Towards_Pathplanning_for_Unmanned_Ground_Vehicles_UGV_in_3D_Plane-Maps_of_Unstructured_Environments},
doi = {10.1007/s13218-011-0098-7},
year = {2011},
date = {2011-01-01},
urldate = {2011-01-01},
journal = {KI - Kuenstliche Intelligenz},
volume = {25},
number = {2},
pages = {141-144},
abstract = {Work in progress on efficient long range path-planning for unmanned ground vehicles (UGV) is presented. It builds upon own work on 3D mapping in unstructured environments, which uses large planar patches for representation and registration of range data. The planar patches allow a very fast assessment of drivability as indicated by experiments with several data-sets.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Work in progress on efficient long range path-planning for unmanned ground vehicles (UGV) is presented. It builds upon own work on 3D mapping in unstructured environments, which uses large planar patches for representation and registration of range data. The planar patches allow a very fast assessment of drivability as indicated by experiments with several data-sets.
2010
Bülow, Heiko; Birk, Andreas; Feyzabadi, Shams
Creating Photo Maps with an Aerial Vehicle in USARsim Book Chapter
In: Baltes, J.; Lagoudakis, M.; Naruse, T.; Shiry, S. (Ed.): RoboCup 2009: Robot WorldCup XIII, vol. 5949, pp. 35-45, Springer, 2010.
@inbook{iFMI-USARsim-RCup10,
title = {Creating Photo Maps with an Aerial Vehicle in USARsim},
author = {Heiko Bülow and Andreas Birk and Shams Feyzabadi},
editor = {J. Baltes and M. Lagoudakis and T. Naruse and S. Shiry},
url = {https://www.researchgate.net/publication/226911605_Creating_Photo_Maps_with_an_Aerial_Vehicle_in_USARsim},
doi = {10.1007/978-3-642-11876-0_4},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {RoboCup 2009: Robot WorldCup XIII},
volume = {5949},
pages = {35-45},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Photo maps, i.e., 2D grids that provide a large scale bird’s eye view of the environment, are of interest for many application scenarios but especially for safety and security missions. We present a very efficient and robust algorithm for this task, which only uses registration between consecutive images, i.e., it does not require any localization. The algorithm is benchmarked in USARsim, where the video stream of a down-looking camera of an aerial vehicle, namely a blimb, is used to generate a large scale photo map.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Photo maps, i.e., 2D grids that provide a large scale bird’s eye view of the environment, are of interest for many application scenarios but especially for safety and security missions. We present a very efficient and robust algorithm for this task, which only uses registration between consecutive images, i.e., it does not require any localization. The algorithm is benchmarked in USARsim, where the video stream of a down-looking camera of an aerial vehicle, namely a blimb, is used to generate a large scale photo map.
Bülow, Heiko; Pfingsthorn, Max; Birk, Andreas
Using Robust Spectral Registration for Scan Matching of Sonar Range Data Proceedings Article
In: 7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC, IFAC, 2010.
@inproceedings{IAV10-Sonar-iFMI,
title = {Using Robust Spectral Registration for Scan Matching of Sonar Range Data},
author = {Heiko Bülow and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/261983604_Using_Robust_Spectral_Registration_for_Scan_Matching_of_Sonar_Range_Data},
doi = {10.3182/20100906-3-IT-2019.00105},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC},
publisher = {IFAC},
abstract = {Scan matching is the process of registering two range scans and is usually applied to laser data. However, such precise sensors are unavailable in underwater environments. Specific algorithms to match sonar range scans have been proposed, but require significant interpretative analysis of the scan to extract range readings similar to laser scanners. This paper presents a fast and robust scan matching method based on spectral registration of rendered scan data. Sonar scans do not need to be pre-processed, other than simple thresholding. The method is tested on a publicly available data set recorded in an abandoned marina.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Scan matching is the process of registering two range scans and is usually applied to laser data. However, such precise sensors are unavailable in underwater environments. Specific algorithms to match sonar range scans have been proposed, but require significant interpretative analysis of the scan to extract range readings similar to laser scanners. This paper presents a fast and robust scan matching method based on spectral registration of rendered scan data. Sonar scans do not need to be pre-processed, other than simple thresholding. The method is tested on a publicly available data set recorded in an abandoned marina.
Birk, Andreas
A Quantitative Assessment of Structural Errors in Grid Maps Journal Article
In: Autonomous Robots, vol. 28, pp. 187-196, 2010.
@article{BrokenMaps-AR10,
title = {A Quantitative Assessment of Structural Errors in Grid Maps},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/220474328_A_quantitative_assessment_of_structural_errors_in_grid_maps},
doi = {10.1007/s10514-009-9159-2},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {Autonomous Robots},
volume = {28},
pages = {187-196},
abstract = {Various common error sources affect the quality of a map, e.g., salt and pepper noise and other forms of noise that are more or less uniformly distributed over the map. But there also exist errors that only occur very rarely in the mapping process but that have severe effects on the final result. They influence not only the local accuracy but also the whole spatial layout of the map. Examples of related error sources include bump noise in the robot’s pose or residual errors in Simultaneous Localization and Mapping (SLAM). The concept of brokenness is introduced in this article to capture the notion of structural errors in grid maps. The map is partitioned into regions that are locally consistent with ground truth but “off” relative to each other. Brokenness measures the number of these regions and their spatial relations. A theoretical basis is introduced to derive the concept of brokenness in a formal way. Furthermore, it is shown how brokenness can be computed in an algorithmic way. Experiments with maps from simulated as well as real world data are presented. They show that the metric can indeed be used to automatically determine the structural quality of a map in a quantitative way.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Various common error sources affect the quality of a map, e.g., salt and pepper noise and other forms of noise that are more or less uniformly distributed over the map. But there also exist errors that only occur very rarely in the mapping process but that have severe effects on the final result. They influence not only the local accuracy but also the whole spatial layout of the map. Examples of related error sources include bump noise in the robot’s pose or residual errors in Simultaneous Localization and Mapping (SLAM). The concept of brokenness is introduced in this article to capture the notion of structural errors in grid maps. The map is partitioned into regions that are locally consistent with ground truth but “off” relative to each other. Brokenness measures the number of these regions and their spatial relations. A theoretical basis is introduced to derive the concept of brokenness in a formal way. Furthermore, it is shown how brokenness can be computed in an algorithmic way. Experiments with maps from simulated as well as real world data are presented. They show that the metric can indeed be used to automatically determine the structural quality of a map in a quantitative way.
Birk, Andreas
Using Recursive Spectral Registrations to Determine Brokenness as Measure of Structural Map Errors Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE, 2010.
@inproceedings{Brokenness-Spectral-ICRA10,
title = {Using Recursive Spectral Registrations to Determine Brokenness as Measure of Structural Map Errors},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/224155860_Using_Recursive_Spectral_Registrations_to_Determine_Brokenness_as_Measure_of_Structural_Map_Errors},
doi = {10.1109/ROBOT.2010.5509322},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE},
abstract = {There are many common error sources that influence mapping, e.g., salt and pepper noise as well as other effects occurring quite uniformly distributed over the map. On the other hand, there are also errors, which occur very rarely but with severe effects. These errors influence not only the local accuracy but the overall spatial layout of the map. Concrete examples include bump noise in the robot's pose or residual errors in Simultaneous Localization and Mapping (SLAM). Brokenness is presented here to capture one form of structural errors in grid maps. Concretely, brokenness measures the degree with which a map can be partitioned into regions that are locally consistent with ground truth but “off” relative to each other. The concept of brokenness is presented in a formal way and it is shown how it can be computed in an efficient way using recursive spectral registrations. Experimental results show that the metric can indeed be used to automatically determine one structural quality aspect of a map in a quantitative way.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
There are many common error sources that influence mapping, e.g., salt and pepper noise as well as other effects occurring quite uniformly distributed over the map. On the other hand, there are also errors, which occur very rarely but with severe effects. These errors influence not only the local accuracy but the overall spatial layout of the map. Concrete examples include bump noise in the robot's pose or residual errors in Simultaneous Localization and Mapping (SLAM). Brokenness is presented here to capture one form of structural errors in grid maps. Concretely, brokenness measures the degree with which a map can be partitioned into regions that are locally consistent with ground truth but “off” relative to each other. The concept of brokenness is presented in a formal way and it is shown how it can be computed in an efficient way using recursive spectral registrations. Experimental results show that the metric can indeed be used to automatically determine one structural quality aspect of a map in a quantitative way.
Birk, Andreas
Where else do you see cheering crowds in a classroom? Journal Article
In: IEEE Robotics and Automation Magazine (RAM), vol. 17, no. 3, pp. 20, 2010.
@article{EduColumn-CheeringCrowds-RAM10,
title = {Where else do you see cheering crowds in a classroom?},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/252059482_Where_Else_Do_You_See_Cheering_Crowds_in_a_Classroom},
doi = {10.1109/MRA.2010.938500},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM)},
volume = {17},
number = {3},
pages = {20},
abstract = {The IEEE Robotics and Automation Society which is an interest group with a wide range of members in industry and academia, including different levels of seniority, application orientation, and experience are discussed. One aspect is that robotics and automation are very interdisciplinary fields. Computer science students are used to an immediate, well-defined effect of an instruction like an assignment of a value to a variable. There is the ultimate hands-on experience of robotics competitions that add a huge, quite special facet to the diverse aspects of robotics and automation education. Although simple, the game had the core ingredients to make robot building and programming challenging and entertaining while teaching important lessons about real-robot systems. Vision-based localization systems had to be developed from scratch in 1997, nowadays excellent software packages are available to start with, especially in the context of education activities.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The IEEE Robotics and Automation Society which is an interest group with a wide range of members in industry and academia, including different levels of seniority, application orientation, and experience are discussed. One aspect is that robotics and automation are very interdisciplinary fields. Computer science students are used to an immediate, well-defined effect of an instruction like an assignment of a value to a variable. There is the ultimate hands-on experience of robotics competitions that add a huge, quite special facet to the diverse aspects of robotics and automation education. Although simple, the game had the core ingredients to make robot building and programming challenging and entertaining while teaching important lessons about real-robot systems. Vision-based localization systems had to be developed from scratch in 1997, nowadays excellent software packages are available to start with, especially in the context of education activities.
Birk, Andreas; Pathak, Kaustubh; Vaskevicius, Narunas; Pfingsthorn, Max; Poppinga, Jann; Schwertfeger, Soeren
Surface Representations for 3D Mapping: A Case for a Paradigm Shift Journal Article
In: KI - German Journal on Artificial Intelligence, 2010.
@article{SurfaceMap-KI-journal2010,
title = {Surface Representations for 3D Mapping: A Case for a Paradigm Shift},
author = {Andreas Birk and Kaustubh Pathak and Narunas Vaskevicius and Max Pfingsthorn and Jann Poppinga and Soeren Schwertfeger},
url = {https://www.researchgate.net/publication/220634405_Surface_Representations_for_3D_Mapping_A_Case_for_a_Paradigm_Shift},
doi = {10.1007/s13218-010-0035-1},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {KI - German Journal on Artificial Intelligence},
abstract = {Point clouds, i.e., sets of 3D coordinates of surface point samples from obstacles, are the predominant form of representation for 3D mapping. They are the raw data format of most 3D sensors and the basis for state of the art algorithms for 3D scan registration. It is argued here that point clouds have severe limitations and a case is made for a necessary paradigm shift to surface based representations. In addition to several conceptual arguments, it is shown how a surface based approach can be used for fast and robust registration of 3D data without the need for robot motion estimates from other sensors. Concretely, a short overview on own work dubbed 3D Plane SLAM is presented. It features an extraction of planes with uncertainties from 3D range scans. Two scans can then be registered by determining the correspondence set that maximizes the global rigid body motion constraint while finding the related optimal decoupled rotations and translations with their underlying uncertainties. The registered scans are embedded in pose-graph SLAM for loop closing and relaxation. },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Point clouds, i.e., sets of 3D coordinates of surface point samples from obstacles, are the predominant form of representation for 3D mapping. They are the raw data format of most 3D sensors and the basis for state of the art algorithms for 3D scan registration. It is argued here that point clouds have severe limitations and a case is made for a necessary paradigm shift to surface based representations. In addition to several conceptual arguments, it is shown how a surface based approach can be used for fast and robust registration of 3D data without the need for robot motion estimates from other sensors. Concretely, a short overview on own work dubbed 3D Plane SLAM is presented. It features an extraction of planes with uncertainties from 3D range scans. Two scans can then be registered by determining the correspondence set that maximizes the global rigid body motion constraint while finding the related optimal decoupled rotations and translations with their underlying uncertainties. The registered scans are embedded in pose-graph SLAM for loop closing and relaxation.
Digor, Elena; Birk, Andreas; Nüchter, Andreas
Exploration Strategies for a Robot with a Continously Rotating 3D Scanner Proceedings Article
In: N., Ando; S., Balakirsky; T., Hemker; M., Reggiani; von Stryk O., (Ed.): Simulation, Modeling, and Programming for Autonomous Robots (SIMPAR), pp. 374-386, Springer, 2010, ISBN: 978-3-642-17319-6.
@inproceedings{3DExploration-ContinuousLRF-Simpar10,
title = {Exploration Strategies for a Robot with a Continously Rotating 3D Scanner},
author = {Elena Digor and Andreas Birk and Andreas Nüchter},
editor = {Ando N. and Balakirsky S. and Hemker T. and Reggiani M. and von Stryk O.},
url = {https://www.researchgate.net/publication/220850158_Exploration_Strategies_for_a_Robot_with_a_Continously_Rotating_3D_Scanner},
doi = {10.1007/978-3-642-17319-6_35},
isbn = {978-3-642-17319-6},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {Simulation, Modeling, and Programming for Autonomous Robots (SIMPAR)},
volume = {6472},
pages = {374-386},
publisher = {Springer},
series = {Lecture Notes in Computer Science (LNCS)},
abstract = {To benchmark the efficiency of exploration strategies one has to use robot simulators. In an exploration task, the robot faces an unknown environment. Of course one could test the algorithm in different real-world scenarios, but a competitive strategy must have good performance in any environment that can be systematically constructed inside a simulator. This paper presents an evaluation of exploration strategies we developed for a specific sensor. A continously rotating 3D laser scanner that scans only into one direction at a time moves through the environment sampling the surrounding. Our evaluation framework features an efficient scanning and robot simulator for kinematic feasible trajectories. We will show that shorter trajectories do not necessarily imply quicker exploration. A simple simulator framework is sufficient for evaluating these properties of path planning algorithms.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
To benchmark the efficiency of exploration strategies one has to use robot simulators. In an exploration task, the robot faces an unknown environment. Of course one could test the algorithm in different real-world scenarios, but a competitive strategy must have good performance in any environment that can be systematically constructed inside a simulator. This paper presents an evaluation of exploration strategies we developed for a specific sensor. A continously rotating 3D laser scanner that scans only into one direction at a time moves through the environment sampling the surrounding. Our evaluation framework features an efficient scanning and robot simulator for kinematic feasible trajectories. We will show that shorter trajectories do not necessarily imply quicker exploration. A simple simulator framework is sufficient for evaluating these properties of path planning algorithms.
Indiveri, G.; Antonelli, G.; Caiti, A.; Casalino, G.; Birk, A.; Pascoal, A.; Caffaz, A.
The CO3-AUVs (Cooperative Cognitive Control for Autonomous Underwater Vehicles) Project: overview and current progresses Proceedings Article
In: 7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC, IFAC, 2010.
@inproceedings{Co3AUVs-IAV10,
title = {The CO3-AUVs (Cooperative Cognitive Control for Autonomous Underwater Vehicles) Project: overview and current progresses},
author = {G. Indiveri and G. Antonelli and A. Caiti and G. Casalino and A. Birk and A. Pascoal and A. Caffaz},
url = {https://www.researchgate.net/publication/269266706_The_CO3AUVs_Cooperative_Cognitive_Control_for_Autonomous_Underwater_Vehicles_Project_overview_and_current_progresses},
doi = {10.3182/20100906-3-IT-2019.00042},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC},
publisher = {IFAC},
abstract = {The paper describes the CO3AUVs - Cooperative Cognitive Control for Autonomous Underwater Vehicles - European Commission Project. This is seventh Framework Programme STREP project under the theme: Information and Communication Technologies Cognitive Systems, Interaction, Robotics. The project started in December 2008. After an overview of the project objectives and consortium composition, the paper describes the project activities.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The paper describes the CO3AUVs - Cooperative Cognitive Control for Autonomous Underwater Vehicles - European Commission Project. This is seventh Framework Programme STREP project under the theme: Information and Communication Technologies Cognitive Systems, Interaction, Robotics. The project started in December 2008. After an overview of the project objectives and consortium composition, the paper describes the project activities.
Pathak, Kaustubh; Birk, Andreas; Vaskevicius, Narunas
Plane-Based Registration of Sonar Data for Underwater 3D Mapping Proceedings Article
In: IEEE International Conference on Intelligent Robots and Systems (IROS), pp. 4880 - 4885, 2010.
@inproceedings{PlaneReg-UnderwaterLesum-IROS2010,
title = {Plane-Based Registration of Sonar Data for Underwater 3D Mapping},
author = {Kaustubh Pathak and Andreas Birk and Narunas Vaskevicius},
url = {https://www.researchgate.net/publication/224200005_Plane-based_registration_of_sonar_data_for_underwater_3D_mapping},
doi = {10.1109/IROS.2010.5650953},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {IEEE International Conference on Intelligent Robots and Systems (IROS)},
pages = {4880 - 4885},
abstract = {Surface-patches based 3D mapping in a real world underwater scenario is presented. It is based on a 6 degrees of freedom registration of sonar data. Planar surfaces are fitted into the sonar data and the subsequent registration method maximizes the overall geometric consistency within a search-space to determine correspondences between the planes. This approach has previously only been used on high quality range data from sensors on land robots like laser range finders. It is shown here that the algorithm is also applicable to very noisy, coarse sonar data. The 3D map presented is of a large underwater structure, namely the Lesumer Sperrwerk, a flood gate north of the city of Bremen, Germany. It is generated from 18 scans collected using a Tritech Eclipse sonar.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Surface-patches based 3D mapping in a real world underwater scenario is presented. It is based on a 6 degrees of freedom registration of sonar data. Planar surfaces are fitted into the sonar data and the subsequent registration method maximizes the overall geometric consistency within a search-space to determine correspondences between the planes. This approach has previously only been used on high quality range data from sensors on land robots like laser range finders. It is shown here that the algorithm is also applicable to very noisy, coarse sonar data. The 3D map presented is of a large underwater structure, namely the Lesumer Sperrwerk, a flood gate north of the city of Bremen, Germany. It is generated from 18 scans collected using a Tritech Eclipse sonar.
Pathak, Kaustubh; Birk, Andreas; Vaskevicius, Narunas; Pfingsthorn, Max; Schwertfeger, Soeren; Poppinga, Jann
Online 3D SLAM by Registration of Large Planar Surface Segments and Closed Form Pose-Graph Relaxation Journal Article
In: Journal of Field Robotics, Special Issue on 3D Mapping, vol. 27, no. 1, pp. 52-84, 2010.
@article{FRJ09-SI-3Dmap-DisasterCity,
title = {Online 3D SLAM by Registration of Large Planar Surface Segments and Closed Form Pose-Graph Relaxation},
author = {Kaustubh Pathak and Andreas Birk and Narunas Vaskevicius and Max Pfingsthorn and Soeren Schwertfeger and Jann Poppinga},
url = {https://www.researchgate.net/publication/220648158_Online_three-dimensional_SLAM_by_registration_of_large_planar_surface_segments_and_closed-form_pose-graph_relaxation},
doi = {10.1002/rob.20322},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {Journal of Field Robotics, Special Issue on 3D Mapping},
volume = {27},
number = {1},
pages = {52-84},
abstract = {A fast pose-graph relaxation technique is presented for enhancing the consistency of three-dimensional (3D) maps created by registering large planar surface patches. The surface patches are extracted from point clouds sampled from a 3D range sensor. The plane-based registration method offers an alternative to the state-of-the-art algorithms and provides advantages in terms of robustness, speed, and storage. One of its features is that it results in an accurate determination of rotation, although a lack of predominant surfaces in certain directions may result in translational uncertainty in those directions. Hence, a loop-closing and relaxation problem is formulated that gains significant speed by relaxing only the translational errors and utilizes the full-translation covariance determined during pairwise registration. This leads to a fast 3D simultaneous localization and mapping suited for online operations. The approach is tested in two disaster scenarios that were mapped at the NIST 2008 Response Robot Evaluation Exercise in Disaster City, Texas. The two data sets from a collapsed car park and a flooding disaster consist of 26 and 70 3D scans, respectively. The results of these experiments show that our approach can generate 3D maps without motion estimates by odometry and that it outperforms iterative closest point–based mapping with respect to speed and robustness.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A fast pose-graph relaxation technique is presented for enhancing the consistency of three-dimensional (3D) maps created by registering large planar surface patches. The surface patches are extracted from point clouds sampled from a 3D range sensor. The plane-based registration method offers an alternative to the state-of-the-art algorithms and provides advantages in terms of robustness, speed, and storage. One of its features is that it results in an accurate determination of rotation, although a lack of predominant surfaces in certain directions may result in translational uncertainty in those directions. Hence, a loop-closing and relaxation problem is formulated that gains significant speed by relaxing only the translational errors and utilizes the full-translation covariance determined during pairwise registration. This leads to a fast 3D simultaneous localization and mapping suited for online operations. The approach is tested in two disaster scenarios that were mapped at the NIST 2008 Response Robot Evaluation Exercise in Disaster City, Texas. The two data sets from a collapsed car park and a flooding disaster consist of 26 and 70 3D scans, respectively. The results of these experiments show that our approach can generate 3D maps without motion estimates by odometry and that it outperforms iterative closest point–based mapping with respect to speed and robustness.
Pathak, Kaustubh; Birk, Andreas; Vaskevicius, Narunas; Poppinga, Jann
Fast Registration Based on Noisy Planes with Unknown Correspondences for 3D Mapping Journal Article
In: IEEE Transactions on Robotics, vol. 26, no. 3, pp. 424-441, 2010.
@article{PlaneMatching-TRO09,
title = {Fast Registration Based on Noisy Planes with Unknown Correspondences for 3D Mapping},
author = {Kaustubh Pathak and Andreas Birk and Narunas Vaskevicius and Jann Poppinga},
url = {https://www.researchgate.net/publication/224124647_Fast_Registration_Based_on_Noisy_Planes_With_Unknown_Correspondences_for_3-D_Mapping},
doi = {10.1109/TRO.2010.2042989},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {IEEE Transactions on Robotics},
volume = {26},
number = {3},
pages = {424-441},
abstract = {We present a robot-pose-registration algorithm, which is entirely based on large planar-surface patches extracted from point clouds sampled from a three-dimensional (3-D) sensor. This approach offers an alternative to the traditional point-to-point iterative-closest-point (ICP) algorithm, its point-to-plane variant, as well as newer grid-based algorithms, such as the 3-D normal distribution transform (NDT). The simpler case of known plane correspondences is tackled first by deriving expressions for least-squares pose estimation considering plane-parameter uncertainty computed during plane extraction. Closed-form expressions for covariances are also derived. To round-off the solution, we present a new algorithm, which is called minimally uncertain maximal consensus (MUMC), to determine the unknown plane correspondences by maximizing geometric consistency by minimizing the uncertainty volume in configuration space. Experimental results from three 3-D sensors, viz., Swiss-Ranger, University of South Florida Odetics Laser Detection and Ranging, and an actuated SICK S300, are given. The first two have low fields of view (FOV) and moderate ranges, while the third has a much bigger FOV and range. Experimental results show that this approach is not only more robust than point- or grid-based approaches in plane-rich environments, but it is also faster, requires significantly less memory, and offers a less-cluttered planar-patches-based visualization.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We present a robot-pose-registration algorithm, which is entirely based on large planar-surface patches extracted from point clouds sampled from a three-dimensional (3-D) sensor. This approach offers an alternative to the traditional point-to-point iterative-closest-point (ICP) algorithm, its point-to-plane variant, as well as newer grid-based algorithms, such as the 3-D normal distribution transform (NDT). The simpler case of known plane correspondences is tackled first by deriving expressions for least-squares pose estimation considering plane-parameter uncertainty computed during plane extraction. Closed-form expressions for covariances are also derived. To round-off the solution, we present a new algorithm, which is called minimally uncertain maximal consensus (MUMC), to determine the unknown plane correspondences by maximizing geometric consistency by minimizing the uncertainty volume in configuration space. Experimental results from three 3-D sensors, viz., Swiss-Ranger, University of South Florida Odetics Laser Detection and Ranging, and an actuated SICK S300, are given. The first two have low fields of view (FOV) and moderate ranges, while the third has a much bigger FOV and range. Experimental results show that this approach is not only more robust than point- or grid-based approaches in plane-rich environments, but it is also faster, requires significantly less memory, and offers a less-cluttered planar-patches-based visualization.
Pathak, Kaustubh; Borrmann, Dorit; Elseberg, Jan; Vaskevicius, Narunas; Birk, Andreas; Nuchter, Andreas
Evaluation of the Robustness of Planar-Patches based 3D-Registration using Marker-based Ground-Truth in an Outdoor Urban Scenario Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), pp. 5725 - 5730, 2010.
@inproceedings{BremenCity-MarkerComparison-IROS2010,
title = {Evaluation of the Robustness of Planar-Patches based 3D-Registration using Marker-based Ground-Truth in an Outdoor Urban Scenario},
author = {Kaustubh Pathak and Dorit Borrmann and Jan Elseberg and Narunas Vaskevicius and Andreas Birk and Andreas Nuchter},
url = {https://www.researchgate.net/publication/224199397_Evaluation_of_the_robustness_of_planar-patches_based_3D-registration_using_marker-based_ground-truth_in_an_outdoor_urban_scenario},
doi = {10.1109/IROS.2010.5649648},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
pages = {5725 - 5730},
abstract = {The recently introduced Minimum Uncertainty Maximum Consensus (MUMC) algorithm for 3D scene registration using planar-patches is tested in a large outdoor urban setting without any prior motion estimate whatsoever. With the aid of a new overlap metric based on unmatched patches, the algorithm is shown to work successfully in most cases. The absolute accuracy of its computed result is corroborated for the first time by ground-truth obtained using reflective markers. There were a couple of unsuccessful scan-pairs. These are analyzed for the reason of failure by formulating two kinds of overlap metrics: one based on the actual overlapping surface-area and another based on the extent of agreement of range-image pixels. We conclude that neither metric in isolation is able to predict all failures, but that both taken together are able to predict the difficulty level of a scan-pair vis-à-vis registration by MUMC.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The recently introduced Minimum Uncertainty Maximum Consensus (MUMC) algorithm for 3D scene registration using planar-patches is tested in a large outdoor urban setting without any prior motion estimate whatsoever. With the aid of a new overlap metric based on unmatched patches, the algorithm is shown to work successfully in most cases. The absolute accuracy of its computed result is corroborated for the first time by ground-truth obtained using reflective markers. There were a couple of unsuccessful scan-pairs. These are analyzed for the reason of failure by formulating two kinds of overlap metrics: one based on the actual overlapping surface-area and another based on the extent of agreement of range-image pixels. We conclude that neither metric in isolation is able to predict all failures, but that both taken together are able to predict the difficulty level of a scan-pair vis-à-vis registration by MUMC.
Pathak, Kaustubh; Vaskevicius, Narunas; Birk, Andreas
Uncertainty Analysis for Optimum Plane Extraction from Noisy 3D Range-Sensor Point-Clouds Journal Article
In: Intelligent Service Robotics, vol. 3, no. 1, pp. 37–48, 2010.
@article{PlaneUncertainty-JISR09,
title = {Uncertainty Analysis for Optimum Plane Extraction from Noisy 3D Range-Sensor Point-Clouds},
author = {Kaustubh Pathak and Narunas Vaskevicius and Andreas Birk},
url = {https://www.researchgate.net/publication/226125925_Uncertainty_Analysis_for_Optimum_Plane_Extraction_from_Noisy_3D_Range-Sensor_Point-Clouds},
doi = {10.1007/s11370-009-0057-4},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {Intelligent Service Robotics},
volume = {3},
number = {1},
pages = {37--48},
abstract = {We utilize a more accurate range noise model for 3D sensors to derive from scratch the expressions for the optimum plane fitting a set of noisy points and for the combined covariance matrix of the plane’s parameters, viz. its normal and its distance to the origin. The range error model used by us is a quadratic function of the true range and also the incidence angle. Closed-form expressions for the Cramér–Rao uncertainty bound are derived and utilized for analyzing four methods of covariance computation: exact maximum likelihood, renormalization, approximate least-squares, and eigenvector perturbation. The effect of the simplifying assumptions inherent in these methods are compared with respect to accuracy, speed, and ease of interpretation of terms. The approximate least-squares covariance matrix is shown to possess a number of desirable properties, e.g., the optimal solution forms its null-space and its components are functions of easily understood terms like the planar-patch’s weighted centroid and scatter. It is also fast to compute and accurate enough in practice. Its experimental application to real-time range-image registration and plane fusion is shown by using a commercially available 3D range sensor.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We utilize a more accurate range noise model for 3D sensors to derive from scratch the expressions for the optimum plane fitting a set of noisy points and for the combined covariance matrix of the plane’s parameters, viz. its normal and its distance to the origin. The range error model used by us is a quadratic function of the true range and also the incidence angle. Closed-form expressions for the Cramér–Rao uncertainty bound are derived and utilized for analyzing four methods of covariance computation: exact maximum likelihood, renormalization, approximate least-squares, and eigenvector perturbation. The effect of the simplifying assumptions inherent in these methods are compared with respect to accuracy, speed, and ease of interpretation of terms. The approximate least-squares covariance matrix is shown to possess a number of desirable properties, e.g., the optimal solution forms its null-space and its components are functions of easily understood terms like the planar-patch’s weighted centroid and scatter. It is also fast to compute and accurate enough in practice. Its experimental application to real-time range-image registration and plane fusion is shown by using a commercially available 3D range sensor.
Pfingsthorn, Max; Schwertfeger, Soeren; Bülow, Heiko; Birk, Andreas
Maximum Likelihood Mapping with Spectral Image Registration Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2010.
@inproceedings{underwater-vSLAM-icra10,
title = {Maximum Likelihood Mapping with Spectral Image Registration},
author = {Max Pfingsthorn and Soeren Schwertfeger and Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/224155898_Maximum_Likelihood_Mapping_with_Spectral_Image_Registration},
doi = {10.1109/ROBOT.2010.5509366},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {A core challenge in probabilistic mapping is to extract meaningful uncertainty information from data registration methods. While this has been investigated in ICP-based scan matching methods, other registration methods have not been analyzed. In this paper, an uncertainty analysis of a Fourier Mellin based image registration algorithm is introduced, which to our knowledge is the first of its kind involving spectral registration. A covariance matrix is extracted from the result of a Phase-Only Matched Filter, which is interpreted as a probability mass function. The method is embedded in a pose graph implementation for Simultaneous Localization and Mapping (SLAM) and validated with experiments in the underwater domain.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A core challenge in probabilistic mapping is to extract meaningful uncertainty information from data registration methods. While this has been investigated in ICP-based scan matching methods, other registration methods have not been analyzed. In this paper, an uncertainty analysis of a Fourier Mellin based image registration algorithm is introduced, which to our knowledge is the first of its kind involving spectral registration. A covariance matrix is extracted from the result of a Phase-Only Matched Filter, which is interpreted as a probability mass function. The method is embedded in a pose graph implementation for Simultaneous Localization and Mapping (SLAM) and validated with experiments in the underwater domain.
Poppinga, Jann; Birk, Andreas; Pathak, Kaustubh
A Characterization of 3D Sensors for Response Robots Book Chapter
In: Baltes, J.; Lagoudakis, M.; Naruse, T.; Shiry, S. (Ed.): RoboCup 2009: Robot WorldCup XIII, vol. 5949, pp. 264-275, Springer, 2010.
@inbook{3DSensorSurvey-Rescue-RCup10,
title = {A Characterization of 3D Sensors for Response Robots},
author = {Jann Poppinga and Andreas Birk and Kaustubh Pathak},
editor = {J. Baltes and M. Lagoudakis and T. Naruse and S. Shiry},
url = {https://www.researchgate.net/publication/220797911_A_Characterization_of_3D_Sensors_for_Response_Robots},
doi = {10.1007/978-3-642-11876-0_23},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {RoboCup 2009: Robot WorldCup XIII},
volume = {5949},
pages = {264-275},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Sensors that measure range information not only in a single plane are becoming more and more important for mobile robots, especially for applications in unstructured environments like response missions where 3D perception and 3D mapping is of interest. Three such sensors are characterized here, namely a Hokuyo URG-04LX laser scanner actuated with a servo in a pitching motion, a Videre STOC stereo camera and a Swissranger SR-3000. The three devices serve as prototypical examples of the according technologies, i.e., 3D laser scanners, stereo vision and time-of-flight cameras.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Sensors that measure range information not only in a single plane are becoming more and more important for mobile robots, especially for applications in unstructured environments like response missions where 3D perception and 3D mapping is of interest. Three such sensors are characterized here, namely a Hokuyo URG-04LX laser scanner actuated with a servo in a pitching motion, a Videre STOC stereo camera and a Swissranger SR-3000. The three devices serve as prototypical examples of the according technologies, i.e., 3D laser scanners, stereo vision and time-of-flight cameras.
Schwertfeger, Soeren; Bülow, Heiko; Birk, Andreas
On the effects of Sampling Resolution in Improved Fourier Mellin based Registration for Underwater Mapping Proceedings Article
In: 7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC, IFAC, 2010.
@inproceedings{iFMI-Resolution-IAV10,
title = {On the effects of Sampling Resolution in Improved Fourier Mellin based Registration for Underwater Mapping},
author = {Soeren Schwertfeger and Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/263258280_On_the_effects_of_Sampling_Resolution_in_Improved_Fourier_Mellin_based_Registration_for_Underwater_Mapping},
doi = {10.3182/20100906-3-IT-2019.00106},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {7th Symposium on Intelligent Autonomous Vehicles (IAV), IFAC},
publisher = {IFAC},
abstract = {The properties of the fast and robust method for visual odometry based on the Fourier-Mellin Invariant (FMI) descriptor are analyzed here. The algorithm is particularly suited for Autonomous Underwater Vehicles (AUV), namely for the online generation of photo maps that can be the basis of intelligent on-board functionalities. The photo maps can be generated just based on registration, i.e., without any information about the vehicle's pose or its motion. The algorithm makes heavy use of 2D Fourier transformations and its speed thus depends on the resolution of the input images and the size of the intermediate matrices used. Using the ground truth path of generated artificial video streams, the effects of different resolutions and other parameters are compared and evaluated with respect to their speed.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The properties of the fast and robust method for visual odometry based on the Fourier-Mellin Invariant (FMI) descriptor are analyzed here. The algorithm is particularly suited for Autonomous Underwater Vehicles (AUV), namely for the online generation of photo maps that can be the basis of intelligent on-board functionalities. The photo maps can be generated just based on registration, i.e., without any information about the vehicle's pose or its motion. The algorithm makes heavy use of 2D Fourier transformations and its speed thus depends on the resolution of the input images and the size of the intermediate matrices used. Using the ground truth path of generated artificial video streams, the effects of different resolutions and other parameters are compared and evaluated with respect to their speed.
Sehgal, Anuj; Cernea, Daniel; Birk, Andreas
Modeling Underwater Acoustic Communications for Multi-Robot Missions in a Robotics Simulator Proceedings Article
In: IEEE Oceans, IEEE Press, 2010.
@inproceedings{UnderwaterCommSimulation-Oceans10,
title = {Modeling Underwater Acoustic Communications for Multi-Robot Missions in a Robotics Simulator},
author = {Anuj Sehgal and Daniel Cernea and Andreas Birk},
url = {https://www.researchgate.net/publication/224183095_Modeling_underwater_acoustic_communications_for_multi-robot_missions_in_a_robotics_simulator},
doi = {10.1109/OCEANSSYD.2010.5603824},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {IEEE Oceans},
publisher = {IEEE Press},
abstract = {Missions involving multiple Autonomous Underwater Vehicles (AUVs) are gaining increasing popularity with the advent of better control mechanisms and availability of acoustic modems that allow for cooperative networked tasks to be carried out by Unmanned Underwater Vehicles (UUVs). However, high costs associated with offshore testing and the lack of useful tools to accurately simulate multi-AUV tasks has been a hindrance to exploiting the full potential of the field. Even though there exist a few simulators for modeling a single AUV, there is virtually no simulation tool available that allows modeling of acoustically networked communications between multiple AUVs by accurately characterizing the underwater acoustic channel. In this paper we present an overview on modeling of the underwater acoustic channel, taking into account the high degree of local variability of ocean conditions, multi-path echoes and ambient noise, within the framework of an underwater acoustic communications server for the Unified System for Automation and Robotics Simulator (USARSim) robotics simulator.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Missions involving multiple Autonomous Underwater Vehicles (AUVs) are gaining increasing popularity with the advent of better control mechanisms and availability of acoustic modems that allow for cooperative networked tasks to be carried out by Unmanned Underwater Vehicles (UUVs). However, high costs associated with offshore testing and the lack of useful tools to accurately simulate multi-AUV tasks has been a hindrance to exploiting the full potential of the field. Even though there exist a few simulators for modeling a single AUV, there is virtually no simulation tool available that allows modeling of acoustically networked communications between multiple AUVs by accurately characterizing the underwater acoustic channel. In this paper we present an overview on modeling of the underwater acoustic channel, taking into account the high degree of local variability of ocean conditions, multi-path echoes and ambient noise, within the framework of an underwater acoustic communications server for the Unified System for Automation and Robotics Simulator (USARSim) robotics simulator.
Vaskevicius, Narunas; Birk, Andreas; Pathak, Kaustubh; Schwertfeger, Sören
Efficient Representation in Three-Dimensional Environment Modeling for Planetary Robotic Exploration Journal Article
In: Advanced Robotics, vol. 24, no. 8-9, pp. 1169-1197, 2010.
@article{ARJ-SpaceRobot-Planes09,
title = {Efficient Representation in Three-Dimensional Environment Modeling for Planetary Robotic Exploration},
author = {Narunas Vaskevicius and Andreas Birk and Kaustubh Pathak and Sören Schwertfeger},
url = {https://www.researchgate.net/publication/220671790_Efficient_Representation_in_Three-Dimensional_Environment_Modeling_for_Planetary_Robotic_Exploration},
doi = {10.1163/016918610X501291},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
journal = {Advanced Robotics},
volume = {24},
number = {8-9},
pages = {1169-1197},
abstract = {Good situation awareness is an absolute must when operating mobile robots for planetary exploration. 3D sensing and modeling data gathered by the robot are hence crucial for the operator. But standard methods based on stereo vision have their limitations, especially in scenarios where there is no or only very limited visibility, e.g., due to extreme light conditions. 3D Laser Range Finders (3D-LRF) provide an interesting alternative, especially as they can provide very accurate, high resolution data at very high sampling rates. But the more 3D range data is acquired, the harder it becomes to transmit the data to the operator station. Here, a fast and robust method to fit planar surface patches into the data is presented. The usefulness of the approach is demonstrated in two different sets of experiments. The first set is based on data from our participation at the ESA Lunar Robotics Challenge 2008. The second one is based on data from a Velodyne 3D-LRF in a high fidelity simulation with ground truth data from Mars.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Good situation awareness is an absolute must when operating mobile robots for planetary exploration. 3D sensing and modeling data gathered by the robot are hence crucial for the operator. But standard methods based on stereo vision have their limitations, especially in scenarios where there is no or only very limited visibility, e.g., due to extreme light conditions. 3D Laser Range Finders (3D-LRF) provide an interesting alternative, especially as they can provide very accurate, high resolution data at very high sampling rates. But the more 3D range data is acquired, the harder it becomes to transmit the data to the operator station. Here, a fast and robust method to fit planar surface patches into the data is presented. The usefulness of the approach is demonstrated in two different sets of experiments. The first set is based on data from our participation at the ESA Lunar Robotics Challenge 2008. The second one is based on data from a Velodyne 3D-LRF in a high fidelity simulation with ground truth data from Mars.
Vaskevicius, Narunas; Pathak, Kaustubh; Pascanu, Razvan; Birk, Andreas
Extraction of Quadrics From Noisy Point-Clouds Using a Sensor Noise Model Proceedings Article
In: IEEE International Conference on Robotics and Automation (ICRA), IEEE Press, 2010.
@inproceedings{Quadrics-Uncertainty-ICRA10,
title = {Extraction of Quadrics From Noisy Point-Clouds Using a Sensor Noise Model},
author = {Narunas Vaskevicius and Kaustubh Pathak and Razvan Pascanu and Andreas Birk},
url = {https://www.researchgate.net/publication/221076381_Extraction_of_Quadrics_From_Noisy_Point-Clouds_Using_a_Sensor_Noise_Model},
doi = {10.1109/ROBOT.2010.5509463},
year = {2010},
date = {2010-01-01},
urldate = {2010-01-01},
booktitle = {IEEE International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE Press},
abstract = {Fitting optimum quadrics on segmented noisy range-image is a challenging task. The algebraic least-squares fit usually considered in the literature is biased, although it is fast to compute. We extend our previous work in planar patch extraction using a detailed sensor noise model to the extraction of quadrics. First, the fast least-squares method is modified to detect degeneracy automatically and it is used to aid segmentation. The final segmented quadric patch is then refined using a numerical maximum likelihood approach which employs the sensor range error model. Experimental results for artificial data and two different 3D sensors are provided to show the feasibility of our approach.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Fitting optimum quadrics on segmented noisy range-image is a challenging task. The algebraic least-squares fit usually considered in the literature is biased, although it is fast to compute. We extend our previous work in planar patch extraction using a detailed sensor noise model to the extraction of quadrics. First, the fast least-squares method is modified to detect degeneracy automatically and it is used to aid segmentation. The final segmented quadric patch is then refined using a numerical maximum likelihood approach which employs the sensor range error model. Experimental results for artificial data and two different 3D sensors are provided to show the feasibility of our approach.
2009
Bülow, Heiko; Birk, Andreas
Fast and Robust Photomapping with an Unmanned Aerial Vehicle (UAV) Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2009.
@inproceedings{iFMI-iros09,
title = {Fast and Robust Photomapping with an Unmanned Aerial Vehicle (UAV)},
author = {Heiko Bülow and Andreas Birk},
url = {https://www.researchgate.net/publication/224090930_Fast_and_Robust_Photomapping_with_an_Unmanned_Aerial_Vehicle_UAV},
doi = {10.1109/IROS.2009.5354505},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {A fast and robust method for visual odometry based on the Fourier-Mellin Invariant (FMI) descriptor is presented. It extends previous FMI based approaches in two ways. First, a logarithmic representation of the spectral magnitude of the FMI descriptor is used. Second, a filter on the frequency where the shift is supposed to appear is applied. It is shown with experiments with an Unmanned Aerial Vehicle that this improved Fourier-Mellin Invariant (iFMI) method is is indeed an advancement and well suited for online visual odometry to generate large photo maps.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A fast and robust method for visual odometry based on the Fourier-Mellin Invariant (FMI) descriptor is presented. It extends previous FMI based approaches in two ways. First, a logarithmic representation of the spectral magnitude of the FMI descriptor is used. Second, a filter on the frequency where the shift is supposed to appear is applied. It is shown with experiments with an Unmanned Aerial Vehicle that this improved Fourier-Mellin Invariant (iFMI) method is is indeed an advancement and well suited for online visual odometry to generate large photo maps.
Bülow, Heiko; Birk, Andreas; Unnithan, Vikram
Online Generation of an Underwater Photo Map with Improved Fourier Mellin based Registration Proceedings Article
In: IEEE OCEANS, IEEE Press, 2009.
@inproceedings{iFMI-UnderwaterPhotoMap-Oceans09,
title = {Online Generation of an Underwater Photo Map with Improved Fourier Mellin based Registration},
author = {Heiko Bülow and Andreas Birk and Vikram Unnithan},
url = {https://www.researchgate.net/publication/224599892_Online_generation_of_an_underwater_photo_map_with_improved_Fourier_Mellin_based_registration},
doi = {10.1109/OCEANSE.2009.5278193},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE OCEANS},
publisher = {IEEE Press},
abstract = {A novel algorithm for stitching images, which is faster and more robust than standard approaches like the Scale Invariant Feature Transform (SIFT) is presented. The algorithm is particularly suited for Autonomous Underwater Vehicles (AUV), namely for the online generation of photo maps that can be the basis of intelligent onboard functionalities. The photo maps can be generated just based on registration, i.e., without any information about the vehicle's pose or its motion. A new method based on the Fourier-Mellin (FM) transform is presented. It is based on a 2D Fourier transform where the shift between two signals is detected by the phase information. In contrast to previous work using FM, a polar-logarithmic resampling of image information is used to turn rotation and scaling into corresponding phase shift and allow for a registration in one step.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A novel algorithm for stitching images, which is faster and more robust than standard approaches like the Scale Invariant Feature Transform (SIFT) is presented. The algorithm is particularly suited for Autonomous Underwater Vehicles (AUV), namely for the online generation of photo maps that can be the basis of intelligent onboard functionalities. The photo maps can be generated just based on registration, i.e., without any information about the vehicle's pose or its motion. A new method based on the Fourier-Mellin (FM) transform is presented. It is based on a 2D Fourier transform where the shift between two signals is detected by the phase information. In contrast to previous work using FM, a polar-logarithmic resampling of image information is used to turn rotation and scaling into corresponding phase shift and allow for a registration in one step.
Birk, Andreas; Antonelli, Gianluca; Pascoal, Antonio; Caffaz, Andrea
Cooperative Cognitive Control for Autonomous Underwater Vehicles Proceedings Article
In: Bertram, Volker (Ed.): 8th International Conference on Computer Applications and Information Technology in the Maritime Industries (COMPIT), 2009.
@inproceedings{co3auvs-compit09,
title = {Cooperative Cognitive Control for Autonomous Underwater Vehicles},
author = {Andreas Birk and Gianluca Antonelli and Antonio Pascoal and Andrea Caffaz},
editor = {Volker Bertram},
url = {https://www.researchgate.net/publication/261990824_Cognitive_Cooperative_Control_for_Autonomous_Underwater_Vehicles},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {8th International Conference on Computer Applications and Information Technology in the Maritime Industries (COMPIT)},
abstract = {This paper describes the plan of work, the methodologies, and the first steps of the EU project Co3-AUV (Cognitive Cooperative Control of AUVs). Started in February 2008, Co3-AUV is led by Jacobs University, Bremen, being IST-Lisbon, ISME (the Italian Interuniversity Research Ctr. on Integrated Systems for the Marine Environment) and GraalTech the other partners. The aim of the Co3-AUVsproject is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects will be investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioural control strictly linked with the underwater communication challenges. As a result, the team of AUVs will cooperate in challenging scenarios in the execution of missions where all data are processed online. In doing so, the team will be robust with respect to failures and environmental changes. These key features will be tested in a harbour scenario where additional difficulties with respect to open sea applications arise and in a human diver assistance scenario that also illustrates human robot interaction issues. },
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the plan of work, the methodologies, and the first steps of the EU project Co3-AUV (Cognitive Cooperative Control of AUVs). Started in February 2008, Co3-AUV is led by Jacobs University, Bremen, being IST-Lisbon, ISME (the Italian Interuniversity Research Ctr. on Integrated Systems for the Marine Environment) and GraalTech the other partners. The aim of the Co3-AUVsproject is to develop, implement and test advanced cognitive systems for coordination and cooperative control of multiple AUVs. Several aspects will be investigated including 3D perception and mapping, cooperative situation awareness, deliberation and navigation as well as behavioural control strictly linked with the underwater communication challenges. As a result, the team of AUVs will cooperate in challenging scenarios in the execution of missions where all data are processed online. In doing so, the team will be robust with respect to failures and environmental changes. These key features will be tested in a harbour scenario where additional difficulties with respect to open sea applications arise and in a human diver assistance scenario that also illustrates human robot interaction issues.
Birk, Andreas; Poppinga, Jann; Pfingsthorn, Max
Using different Humanoid Robots for Science Edutainment of Secondary School Pupils Book Chapter
In: Iocchi, Luca; Matsubara, Hitoshi; Weitzenfeld, Alfredo; Zhou, Changjiu (Ed.): RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI), vol. 5399, pp. 451-462, Springer, 2009.
@inbook{InnoCampHumanoid-RoboCup08,
title = {Using different Humanoid Robots for Science Edutainment of Secondary School Pupils},
author = {Andreas Birk and Jann Poppinga and Max Pfingsthorn},
editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
url = {https://www.researchgate.net/publication/220797318_Using_Different_Humanoid_Robots_for_Science_Edutainment_of_Secondary_School_Pupils},
doi = {10.1007/978-3-642-02921-9_39},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI)},
volume = {5399},
pages = {451-462},
publisher = {Springer},
abstract = {Robotics camps that involve design, construction and programming tasks are a popular part of various educational activities. This paper presents the results of a survey that accompanied the Innovationscamp, a one week intensive workshop for promoting science and engineering among secondary school pupils through humanoid robots. Two very different types of platforms were used in this workshop: LEGO mindstorms, which are widely used for educational activities, and Bioloid humanoids, which are more commonly used for professional research. Though the workshop participants were robotics novices, the survey indicates through several statistically significant results that the Bioloid robots are preferred by the pupils over the LEGO robots as educational tools.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Robotics camps that involve design, construction and programming tasks are a popular part of various educational activities. This paper presents the results of a survey that accompanied the Innovationscamp, a one week intensive workshop for promoting science and engineering among secondary school pupils through humanoid robots. Two very different types of platforms were used in this workshop: LEGO mindstorms, which are widely used for educational activities, and Bioloid humanoids, which are more commonly used for professional research. Though the workshop participants were robotics novices, the survey indicates through several statistically significant results that the Bioloid robots are preferred by the pupils over the LEGO robots as educational tools.
Birk, Andreas; Poppinga, Jann; Stoyanov, Todor; Nevatia, Yashodhan
Planetary Exploration in USARsim: A Case Study including Real World Data from Mars Book Chapter
In: Iocchi, Luca; Matsubara, Hitoshi; Weitzenfeld, Alfredo; Zhou, Changjiu (Ed.): RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI), vol. 5399, pp. 463-472, Springer, 2009.
@inbook{VirtualMars-RoboCup08,
title = {Planetary Exploration in USARsim: A Case Study including Real World Data from Mars},
author = {Andreas Birk and Jann Poppinga and Todor Stoyanov and Yashodhan Nevatia},
editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
url = {https://www.researchgate.net/publication/220797860_Planetary_Exploration_in_USARsim_A_Case_Study_Including_Real_World_Data_from_Mars},
doi = {10.1007/978-3-642-02921-9_40},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI)},
volume = {5399},
pages = {463-472},
publisher = {Springer},
abstract = {Intelligent Mobile Robots are increasingly used in unstructured domains; one particularly challenging example for this is planetary exploration. The preparation of according missions is highly non-trivial, especially as it is difficult to carry out realistic experiments without very sophisticated infrastructures. In this paper, we argue that the Unified System for Automation and Robot Simulation (USARSim) offers interesting opportunities for research on planetary exploration by mobile robots. With the example of work on terrain classification, it is shown how synthetic as well as real world data from Mars can be used to test an algorithm’s performance in USARSim. Concretely, experiments with an algorithm for the detection of negotiable ground on a planetary surface are presented. It is shown that the approach performs fast and robust on planetary surfaces.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Intelligent Mobile Robots are increasingly used in unstructured domains; one particularly challenging example for this is planetary exploration. The preparation of according missions is highly non-trivial, especially as it is difficult to carry out realistic experiments without very sophisticated infrastructures. In this paper, we argue that the Unified System for Automation and Robot Simulation (USARSim) offers interesting opportunities for research on planetary exploration by mobile robots. With the example of work on terrain classification, it is shown how synthetic as well as real world data from Mars can be used to test an algorithm’s performance in USARSim. Concretely, experiments with an algorithm for the detection of negotiable ground on a planetary surface are presented. It is shown that the approach performs fast and robust on planetary surfaces.
Birk, Andreas; Schwertfeger, Soeren; Pathak, Kaustubh
3D Data Collection at Disaster City at the 2008 NIST Response Robot Evaluation Exercise (RREE) Proceedings Article
In: International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2009.
@inproceedings{SSRR09-RREE08-DataCollection,
title = {3D Data Collection at Disaster City at the 2008 NIST Response Robot Evaluation Exercise (RREE)},
author = {Andreas Birk and Soeren Schwertfeger and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/240645222_3D_data_collection_at_disaster_city_at_the_2008_NIST_Response_Robot_Evaluation_Exercise_RREE},
doi = {10.1109/SSRR.2009.5424164},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {A collection of 3D data sets gathered at the 2008 NIST Response Robot Evaluation Exercise (RREE) in Disaster City, Texas is described. The data sets consist of 3D point clouds collected with an actuated laser range finder in different disaster scenarios. The data sets can be used for performance evaluation of robotics algorithms, especially for 3D mapping. An example is discussed where a 3D model is generated from scans taken in a collapsed car parking.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A collection of 3D data sets gathered at the 2008 NIST Response Robot Evaluation Exercise (RREE) in Disaster City, Texas is described. The data sets consist of 3D point clouds collected with an actuated laser range finder in different disaster scenarios. The data sets can be used for performance evaluation of robotics algorithms, especially for 3D mapping. An example is discussed where a 3D model is generated from scans taken in a collapsed car parking.
Birk, Andreas; Schwertfeger, Soeren; Pathak, Kaustubh
A Networking Framework for Teleoperation in Safety, Security, and Rescue Robotics Journal Article
In: IEEE Wireless Communications, vol. 6, no. 1, pp. 6-13, 2009.
@article{NetworkingFramework-WirelessRobots09,
title = {A Networking Framework for Teleoperation in Safety, Security, and Rescue Robotics},
author = {Andreas Birk and Soeren Schwertfeger and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/224396377_A_networking_framework_for_teleoperation_in_safety_security_and_rescue_robotics},
doi = {10.1109/MWC.2009.4804363},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
journal = {IEEE Wireless Communications},
volume = {6},
number = {1},
pages = {6-13},
abstract = {Safety, security, and rescue robotics is an important application field that can be viewed as a prototypical example of a domain where networked mobile robots are used for the exploration of unstructured environments that are inaccessible to or dangerous for humans. Teleoperation, based on wireless networks, is much more complex than what one might expect at first glance because it goes well beyond mere mappings of low-level user inputs - like joystick commands - to motor activations on a robot. Teleoperation for SSRR must move up to the behavior and mission levels where a single operator triggers short-time, autonomous behaviors, respectively, and supervises a whole team of autonomously operating robots. Consequently, a significant amount of heterogeneous data - video, maps, goal points, victim data, and so on - must be transmitted between robots and mission control. In this article, a networking framework for teleoperation in SSRR is presented. It was evaluated in a series of field tests and competitions, including the European Land Robot Trials and RoboCup events.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Safety, security, and rescue robotics is an important application field that can be viewed as a prototypical example of a domain where networked mobile robots are used for the exploration of unstructured environments that are inaccessible to or dangerous for humans. Teleoperation, based on wireless networks, is much more complex than what one might expect at first glance because it goes well beyond mere mappings of low-level user inputs - like joystick commands - to motor activations on a robot. Teleoperation for SSRR must move up to the behavior and mission levels where a single operator triggers short-time, autonomous behaviors, respectively, and supervises a whole team of autonomously operating robots. Consequently, a significant amount of heterogeneous data - video, maps, goal points, victim data, and so on - must be transmitted between robots and mission control. In this article, a networking framework for teleoperation in SSRR is presented. It was evaluated in a series of field tests and competitions, including the European Land Robot Trials and RoboCup events.
Birk, Andreas; Vaskevicius, Narunas; Pathak, Kaustubh; Schwertfeger, Soeren; Poppinga, Jann; Bülow, Heiko
3-D Perception and Modeling Journal Article
In: IEEE Robotics and Automation Magazine (RAM), special issue on Space Robotics, R. Wagner, R. Volpe, G. Visentin (Eds), vol. 6, no. 4, pp. 53-60, 2009.
@article{ESA-LRC-3D-RAM09,
title = {3-D Perception and Modeling},
author = {Andreas Birk and Narunas Vaskevicius and Kaustubh Pathak and Soeren Schwertfeger and Jann Poppinga and Heiko Bülow},
url = {https://www.researchgate.net/publication/224611227_3-D_perception_and_modeling},
doi = {10.1109/MRA.2009.934822},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
journal = {IEEE Robotics and Automation Magazine (RAM), special issue on Space Robotics, R. Wagner, R. Volpe, G. Visentin (Eds)},
volume = {6},
number = {4},
pages = {53-60},
abstract = {In the context of the 2008 Lunar Robotics Challenge (LRC) of the European Space Agency (ESA), the Jacobs Robotics team investigated three-dimensional (3-D) perception and modeling as an important basis of autonomy in unstructured domains. Concretely, the efficient modeling of the terrain via a 3D laser range finder (LRF) is addressed. The underlying fast extraction of planar surface patches can be used to improve situational awareness of an operator or for path planning. 3D perception and modeling is an important basis for mobile robot operations in planetary exploration scenarios as it supports good situation awareness for motion level teleoperation as well as higher level intelligent autonomous functions. It is hence desirable to get long-range 3D data with high resolution, large field of view, and very fast update rates. 3D LRF have a high potential in this respect. In addition, 3D LRF can operate under conditions where standard vision based methods fail, e.g., under extreme light conditions. However, it is nontrivial to transmit the huge amount of data delivered by a 3D LRF to an operator station or to use this point cloud data as basis for higher level intelligent functions. Based on our participation in the LRC of the ESA, it is shown how the huge amount of 3D point cloud data from 3D LRF can be tremendously reduced. Concretely, large sets of points are replaced by planar surface patches that are fitted into the data in an optimal way. The underlying computations are very efficient and hence suited for online computations onboard of the robot.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In the context of the 2008 Lunar Robotics Challenge (LRC) of the European Space Agency (ESA), the Jacobs Robotics team investigated three-dimensional (3-D) perception and modeling as an important basis of autonomy in unstructured domains. Concretely, the efficient modeling of the terrain via a 3D laser range finder (LRF) is addressed. The underlying fast extraction of planar surface patches can be used to improve situational awareness of an operator or for path planning. 3D perception and modeling is an important basis for mobile robot operations in planetary exploration scenarios as it supports good situation awareness for motion level teleoperation as well as higher level intelligent autonomous functions. It is hence desirable to get long-range 3D data with high resolution, large field of view, and very fast update rates. 3D LRF have a high potential in this respect. In addition, 3D LRF can operate under conditions where standard vision based methods fail, e.g., under extreme light conditions. However, it is nontrivial to transmit the huge amount of data delivered by a 3D LRF to an operator station or to use this point cloud data as basis for higher level intelligent functions. Based on our participation in the LRC of the ESA, it is shown how the huge amount of 3D point cloud data from 3D LRF can be tremendously reduced. Concretely, large sets of points are replaced by planar surface patches that are fitted into the data in an optimal way. The underlying computations are very efficient and hence suited for online computations onboard of the robot.
Birk, Andreas; Wiggerich, Burkhard; Bülow, Heiko; Pfingsthorn, Max; Schwertfeger, Soeren
Reconnaissance and Camp Security Missions with an Unmanned Aerial Vehicle (UAV) at the 2009 European Land Robots Trials (ELROB) Proceedings Article
In: IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2009.
@inproceedings{SSRR09-ELROB-UAV,
title = {Reconnaissance and Camp Security Missions with an Unmanned Aerial Vehicle (UAV) at the 2009 European Land Robots Trials (ELROB)},
author = {Andreas Birk and Burkhard Wiggerich and Heiko Bülow and Max Pfingsthorn and Soeren Schwertfeger},
url = {https://www.researchgate.net/publication/224120714_Reconnaissance_and_camp_security_missions_with_an_Unmanned_Aerial_Vehicle_UAV_at_the_2009_European_Land_Robots_Trials_ELROB},
doi = {10.1109/SSRR.2009.5424163},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {Results from two response missions at the 2009 European Land Robot Trials (ELROB) are presented. An unmanned aerial vehicle (UAV) in form of an Airrobot AR100-B is used in a reconnaissance and in a camp security scenario. The UAV is capable of autonomous waypoint navigation using onboard GPS processing. A digital video stream from the vehicle is used to create photo maps - also known as mosaicking - in real time at the operator station. This mapping is done using an enhanced version of Fourier Mellin based registration, which turns out to be very fast and robust.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Results from two response missions at the 2009 European Land Robot Trials (ELROB) are presented. An unmanned aerial vehicle (UAV) in form of an Airrobot AR100-B is used in a reconnaissance and in a camp security scenario. The UAV is capable of autonomous waypoint navigation using onboard GPS processing. A digital video stream from the vehicle is used to create photo maps - also known as mosaicking - in real time at the operator station. This mapping is done using an enhanced version of Fourier Mellin based registration, which turns out to be very fast and robust.
Bastani, Hamed; Birk, Andreas
Iterative Localization Optimization for Networked Mobile Robots Proceedings Article
In: IEEE 14th International Conference on Advanced Robotics (ICAR), IEEE Press, 2009.
@inproceedings{RF-Localization-RoleManagement-ICAR09,
title = {Iterative Localization Optimization for Networked Mobile Robots},
author = {Hamed Bastani and Andreas Birk },
url = {https://www.researchgate.net/publication/251892134_Iterative_localization_optimization_for_networked_mobile_robots},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE 14th International Conference on Advanced Robotics (ICAR)},
publisher = {IEEE Press},
abstract = {In an ad-hoc wireless network with mobile nodes, we consider communication links, which are not only used for data transfer, but which also are have the capability of link-length-measurements. We propose a method for localizing the whole network without the need of any anchor or a priori reference information. In other words, the network uses itself as an infrastructure for localization. It is shown how the infrastructure can dynamically move to build a larger covered area and minimize the positioning error. For a considerable population of nodes, it matters how proper constrains are utilized to efficiently keep the nodes within the communication range, as well as a robustly bounding the accumulative positioning error. In other words, the strategy is desired to be able to afford nodes' movements drawbacks. Extensive amount of simulations with promising results are analytically presented here, to evaluate the performance and robustness of the proposed algorithm against environmental noise and positioning inaccuracy.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In an ad-hoc wireless network with mobile nodes, we consider communication links, which are not only used for data transfer, but which also are have the capability of link-length-measurements. We propose a method for localizing the whole network without the need of any anchor or a priori reference information. In other words, the network uses itself as an infrastructure for localization. It is shown how the infrastructure can dynamically move to build a larger covered area and minimize the positioning error. For a considerable population of nodes, it matters how proper constrains are utilized to efficiently keep the nodes within the communication range, as well as a robustly bounding the accumulative positioning error. In other words, the strategy is desired to be able to afford nodes' movements drawbacks. Extensive amount of simulations with promising results are analytically presented here, to evaluate the performance and robustness of the proposed algorithm against environmental noise and positioning inaccuracy.
Birk, Hamed Bastani; Andreas,
Precise Realtime Localization by RF Transceiver ToF Measurements Proceedings Article
In: IEEE 14th International Conference on Advanced Robotics (ICAR), IEEE Press, 2009.
@inproceedings{RF-Localization-MDS-ICAR09,
title = {Precise Realtime Localization by RF Transceiver ToF Measurements},
author = {Hamed Bastani Birk and Andreas},
url = {https://www.researchgate.net/publication/224567942_Precise_realtime_localization_by_RF_transceiver_ToF_measurements},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE 14th International Conference on Advanced Robotics (ICAR)},
publisher = {IEEE Press},
abstract = {A positioning system is presented, which is based on wireless transceivers capable of measuring time of flight. As raw measurements still suffer from distortion and environmentally caused noise, mathematical tools are employed to significantly improve the quality and accuracy of positioning in realtime. Implementation results and simulation of larger scale networks show a very promising bounded error limit of positioning.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A positioning system is presented, which is based on wireless transceivers capable of measuring time of flight. As raw measurements still suffer from distortion and environmentally caused noise, mathematical tools are employed to significantly improve the quality and accuracy of positioning in realtime. Implementation results and simulation of larger scale networks show a very promising bounded error limit of positioning.
Pathak, K.; Pfingsthorn, M.; Vaskevicius, N.; Birk, A.
Relaxing loop-closing errors in 3D maps based on planar surface patches Proceedings Article
In: IEEE International Conference on Advanced Robotics (ICAR), pp. 1-6, 2009.
@inproceedings{PlaneSLAM-LoopClosing-ICAR09,
title = {Relaxing loop-closing errors in 3D maps based on planar surface patches},
author = {K. Pathak and M. Pfingsthorn and N. Vaskevicius and A. Birk},
url = {https://www.researchgate.net/publication/251892149_Relaxing_loop-closing_errors_in_3D_maps_based_on_planar_surface_patches},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE International Conference on Advanced Robotics (ICAR)},
pages = {1-6},
abstract = {3D mapping using large planar patches results in a compact and easily understandable representation of the environment as compared to point-cloud or voxel based representations. In this work, relaxation of loop-closing errors in a 3D map is presented. It is based on a plane-matching algorithm for correspondence-finding and least-squares registration of large planar patches fitted on 3D range-images. This method also provides covariance matrices for the pose-registration result. Based on this registration algorithm, a relaxation method utilizing these covariances is formulated. In particular, we exploit the plane-matcher properties that it provides an accurate rotation estimate, and that it can easily identify principal translational directions of uncertainty, to relax only the translation errors. This results in a closed-form solution that can be computed in a very fast manner, namely within a few milliseconds, as demonstrated by experimental results in an indoor locomotion test arena in form of a high bay rack.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
3D mapping using large planar patches results in a compact and easily understandable representation of the environment as compared to point-cloud or voxel based representations. In this work, relaxation of loop-closing errors in a 3D map is presented. It is based on a plane-matching algorithm for correspondence-finding and least-squares registration of large planar patches fitted on 3D range-images. This method also provides covariance matrices for the pose-registration result. Based on this registration algorithm, a relaxation method utilizing these covariances is formulated. In particular, we exploit the plane-matcher properties that it provides an accurate rotation estimate, and that it can easily identify principal translational directions of uncertainty, to relax only the translation errors. This results in a closed-form solution that can be computed in a very fast manner, namely within a few milliseconds, as demonstrated by experimental results in an indoor locomotion test arena in form of a high bay rack.
Pathak, Kaustubh; Vaskevicius, Narunas; Birk, Andreas
Revisiting Uncertainty Analysis for Optimum Planes Extracted from 3D Range Sensor Point-Clouds Proceedings Article
In: International Conference on Robotics and Automation (ICRA), pp. 1631 - 1636, IEEE press, 2009.
@inproceedings{plane-uncertainty-icra09,
title = {Revisiting Uncertainty Analysis for Optimum Planes Extracted from 3D Range Sensor Point-Clouds},
author = {Kaustubh Pathak and Narunas Vaskevicius and Andreas Birk},
url = {https://www.researchgate.net/publication/224557283_Revisiting_uncertainty_analysis_for_optimum_planes_extracted_from_3D_range_sensor_point-clouds},
doi = {10.1109/ROBOT.2009.5152502},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {International Conference on Robotics and Automation (ICRA)},
pages = {1631 - 1636},
publisher = {IEEE press},
abstract = {In this work, we utilize a recently studied more accurate range noise model for 3D sensors to derive from scratch the expressions for the optimum plane which best fits a point-cloud and for the combined covariance matrix of the plane's parameters. The parameters in question are the plane's normal and its distance to the origin. The range standard-deviation model used by us is a quadratic function of the true range and is a function of the incidence angle as well. We show that for this model, the maximum-likelihood plane is biased, whereas the least-squares plane is not. The plane-parameters' covariance matrix for the least-squares plane is shown to possess a number of desirable properties, e.g., the optimal solution forms its null-space and its components are functions of easily understood terms like the planar-patch's center and scatter. We verify our covariance expression with that obtained by the eigenvector perturbation method. We further compare our method to that of renormalization with respect to the theoretically best covariance matrix in simulation. The application of our approach to real-time range-image registration and plane fusion is shown by an example using a commercially available 3D range sensor. Results show that our method has good accuracy, is fast to compute, and is easy to interpret intuitively.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
In this work, we utilize a recently studied more accurate range noise model for 3D sensors to derive from scratch the expressions for the optimum plane which best fits a point-cloud and for the combined covariance matrix of the plane's parameters. The parameters in question are the plane's normal and its distance to the origin. The range standard-deviation model used by us is a quadratic function of the true range and is a function of the incidence angle as well. We show that for this model, the maximum-likelihood plane is biased, whereas the least-squares plane is not. The plane-parameters' covariance matrix for the least-squares plane is shown to possess a number of desirable properties, e.g., the optimal solution forms its null-space and its components are functions of easily understood terms like the planar-patch's center and scatter. We verify our covariance expression with that obtained by the eigenvector perturbation method. We further compare our method to that of renormalization with respect to the theoretically best covariance matrix in simulation. The application of our approach to real-time range-image registration and plane fusion is shown by an example using a commercially available 3D range sensor. Results show that our method has good accuracy, is fast to compute, and is easy to interpret intuitively.
Pathak, Kaustubh; Vaskevicius, Narunas; Poppinga, Jann; Pfingsthorn, Max; Schwertfeger, Soeren; Birk, Andreas
Fast 3D Mapping by Matching Planes Extracted from Range Sensor Point-Clouds Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2009.
@inproceedings{3Dmap-IROS09,
title = {Fast 3D Mapping by Matching Planes Extracted from Range Sensor Point-Clouds},
author = {Kaustubh Pathak and Narunas Vaskevicius and Jann Poppinga and Max Pfingsthorn and Soeren Schwertfeger and Andreas Birk},
url = {https://www.researchgate.net/publication/224090528_Fast_3D_Mapping_by_Matching_Planes_Extracted_from_Range_Sensor_Point-Clouds},
doi = {10.1109/IROS.2009.5354061},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {This article addresses fast 3D mapping by a mobile robot in a predominantly planar environment. It is based on a novel pose registration algorithm based entirely on matching features composed of plane-segments extracted from point-clouds sampled from a 3D sensor. The approach has advantages in terms of robustness, speed and storage as compared to the voxel based approaches. Unlike previous approaches, the uncertainty in plane parameters is utilized to compute the uncertainty in the pose computed by scan-registration. The algorithm is illustrated by creating a full 3D model of a multi-level robot testing arena.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This article addresses fast 3D mapping by a mobile robot in a predominantly planar environment. It is based on a novel pose registration algorithm based entirely on matching features composed of plane-segments extracted from point-clouds sampled from a 3D sensor. The approach has advantages in terms of robustness, speed and storage as compared to the voxel based approaches. Unlike previous approaches, the uncertainty in plane parameters is utilized to compute the uncertainty in the pose computed by scan-registration. The algorithm is illustrated by creating a full 3D model of a multi-level robot testing arena.
Pfingsthorn, Max; Nevatia, Yashodhan; Stoyanov, Todor; Rathnam, Ravi; Markov, Stefan; Birk, Andreas
Towards Collaborative and Decentralized Mapping in the Jacobs Virtual Rescue Team Book Chapter
In: Iocchi, Luca; Matsubara, Hitoshi; Weitzenfeld, Alfredo; Zhou, Changjiu (Ed.): RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI), vol. 5399, pp. 225-234, Springer, 2009.
@inbook{VirtualMapping-RoboCup08,
title = {Towards Collaborative and Decentralized Mapping in the Jacobs Virtual Rescue Team},
author = {Max Pfingsthorn and Yashodhan Nevatia and Todor Stoyanov and Ravi Rathnam and Stefan Markov and Andreas Birk},
editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
url = {https://www.researchgate.net/publication/220798012_Towards_Cooperative_and_Decentralized_Mapping_in_the_Jacobs_Virtual_Rescue_Team},
doi = {10.1007/978-3-642-02921-9_20},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI)},
volume = {5399},
pages = {225-234},
publisher = {Springer},
abstract = {The task of mapping and exploring an unknown environment remains one of the fundamental problems of mobile robotics. It is a task that can intuitively benefit significantly from a multi-robot approach. In this paper, we describe the design of the multi-robot mapping system used in the Jacobs Virtual Rescue team. The team competed in the World Cup 2007 and won the second place. It is shown how the recently proposed pose graph map representation facilitates not only map merging but also allows transmitting map updates efficiently.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The task of mapping and exploring an unknown environment remains one of the fundamental problems of mobile robotics. It is a task that can intuitively benefit significantly from a multi-robot approach. In this paper, we describe the design of the multi-robot mapping system used in the Jacobs Virtual Rescue team. The team competed in the World Cup 2007 and won the second place. It is shown how the recently proposed pose graph map representation facilitates not only map merging but also allows transmitting map updates efficiently.
Poppinga, Jann; Birk, Andreas
A Novel Approach to Wrap Around Error Correction for a Time-Of-Flight 3D Camera Book Chapter
In: Iocchi, Luca; Matsubara, Hitoshi; Weitzenfeld, Alfredo; Zhou, Changjiu (Ed.): RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI), vol. 5399, pp. 247-258, Springer, 2009.
@inbook{SwissrangerWrapAroundError-RoboCup08,
title = {A Novel Approach to Wrap Around Error Correction for a Time-Of-Flight 3D Camera},
author = {Jann Poppinga and Andreas Birk},
editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
url = {https://www.researchgate.net/publication/220797546_A_Novel_Approach_to_Efficient_Error_Correction_for_the_SwissRanger_Time-of-Flight_3D_Camera},
doi = {10.1007/978-3-642-02921-9_22},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI)},
volume = {5399},
pages = {247-258},
publisher = {Springer},
abstract = {3D data acquisition gets increasingly important for mobile robotics in general and for Safety, Security and Rescue Robotics (SSRR) in particular. 3D data allows for example to estimate the size of gaps, to construct realistic maps of unstructured disaster environments, or to detect human victims from shape. The SwissRanger SR-3000 time-of-flight 3D camera is a popular device for acquiring 3D range data as it offers the fast update rates of a camera with a resolution of 176 × 144 pixel at a reasonable cost. But the SR-3000 suffers - like any device using phase differences of modulated light - from the fundamental problem of wrap around error, i.e., distances that are a multiple of the wavelength of the modulated ranging signal can not be distinguished. The standard solution to this problem is to use an amplitude threshold, i.e., to discard pixels which are relatively dark and hence assumed to be far away. Here, a significant improvement to the standard method is presented that relates measured brightness and distance and also takes the geometry of the modulated light source into account. It is shown that a significantly higher amount of valid range data can be acquired with this new method.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
3D data acquisition gets increasingly important for mobile robotics in general and for Safety, Security and Rescue Robotics (SSRR) in particular. 3D data allows for example to estimate the size of gaps, to construct realistic maps of unstructured disaster environments, or to detect human victims from shape. The SwissRanger SR-3000 time-of-flight 3D camera is a popular device for acquiring 3D range data as it offers the fast update rates of a camera with a resolution of 176 × 144 pixel at a reasonable cost. But the SR-3000 suffers - like any device using phase differences of modulated light - from the fundamental problem of wrap around error, i.e., distances that are a multiple of the wavelength of the modulated ranging signal can not be distinguished. The standard solution to this problem is to use an amplitude threshold, i.e., to discard pixels which are relatively dark and hence assumed to be far away. Here, a significant improvement to the standard method is presented that relates measured brightness and distance and also takes the geometry of the modulated light source into account. It is shown that a significantly higher amount of valid range data can be acquired with this new method.
Rathnam, Ravi; Pfingsthorn, Max; Birk, Andreas
Incorporating large scale SSRR Scenarios into the High Fidelity Simulator USARSim Proceedings Article
In: IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR), pp. 1-6, IEEE Press, 2009.
@inproceedings{SSRR09-ELROB-USARsim,
title = {Incorporating large scale SSRR Scenarios into the High Fidelity Simulator USARSim},
author = {Ravi Rathnam and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/240645221_Incorporating_large_scale_SSRR_scenarios_into_the_high_fidelity_simulator_USARSim},
doi = {10.1109/SSRR.2009.5424161},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
pages = {1-6},
publisher = {IEEE Press},
abstract = {The Unified System for Automation and Robot Simulation is a very popular robot simulator based on the commercial Unreal game engine by Epic Games. Tools shipped with Unreal Tournament 2004 allow to quickly put together simple simulation scenarios. The simulator is particularly suited for research on safety, security, and rescue robotics (SSRR) as demonstrated by its usage by various research groups and as basis for the RoboCup Rescue Virtual League. However, even through its use in international competitions like RoboCup, only a small number of high quality and large scale simulation worlds have been designed. This paper presents a collection of tools developed at Jacobs University Bremen to mitigate this situation and allow the conversion and import of large scale virtual environments in a variety of industry leading 3D modeling formats. Results are presented based on the conversion of a large environment model of a German military base and an adjacent town.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The Unified System for Automation and Robot Simulation is a very popular robot simulator based on the commercial Unreal game engine by Epic Games. Tools shipped with Unreal Tournament 2004 allow to quickly put together simple simulation scenarios. The simulator is particularly suited for research on safety, security, and rescue robotics (SSRR) as demonstrated by its usage by various research groups and as basis for the RoboCup Rescue Virtual League. However, even through its use in international competitions like RoboCup, only a small number of high quality and large scale simulation worlds have been designed. This paper presents a collection of tools developed at Jacobs University Bremen to mitigate this situation and allow the conversion and import of large scale virtual environments in a variety of industry leading 3D modeling formats. Results are presented based on the conversion of a large environment model of a German military base and an adjacent town.
Varsadan, Ioana; Birk, Andreas; Pfingsthorn, Max
Determining Map Quality through an Image Similarity Metric Book Chapter
In: Iocchi, Luca; Matsubara, Hitoshi; Weitzenfeld, Alfredo; Zhou, Changjiu (Ed.): RoboCup 2008: Robot WorldCup XII, Lecture Notes in Artificial Intelligence (LNAI), pp. 355-365, Springer, 2009.
@inbook{MapQuality-RoboCup08,
title = {Determining Map Quality through an Image Similarity Metric},
author = {Ioana Varsadan and Andreas Birk and Max Pfingsthorn},
editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
url = {https://www.researchgate.net/publication/220797769_Determining_Map_Quality_through_an_Image_Similarity_Metric},
doi = {10.1007/978-3-642-02921-9_31},
year = {2009},
date = {2009-01-01},
urldate = {2009-01-01},
booktitle = {RoboCup 2008: Robot WorldCup XII,
Lecture Notes in Artificial Intelligence (LNAI)},
pages = {355-365},
publisher = {Springer},
abstract = {A quantitative assessment of the quality of a robot generated map is of high interest for many reasons. First of all, it allows individual researchers to quantify the quality of their mapping approach and to study the effects of system specific choices like different parameter values in an objective way. Second, it allows peer groups to rank the quality of their different approaches to determine scientific progress; similarly, it allows rankings within competition environments like RoboCup. A quantitative assessment of map quality based on an image similarity metric Ψ is introduced here. It is shown through synthetic as well as through real world data that the metric captures intuitive notions of map quality. Furthermore, the metric is compared to a seemingly more straightforward metric based on Least Mean Squared Euclidean distances (LMS-ED) between map points and ground truth. It is shown that both capture intuitive notions of map quality in a similar way, but that Ψ can be computed much more efficiently than the LMS-ED.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
A quantitative assessment of the quality of a robot generated map is of high interest for many reasons. First of all, it allows individual researchers to quantify the quality of their mapping approach and to study the effects of system specific choices like different parameter values in an objective way. Second, it allows peer groups to rank the quality of their different approaches to determine scientific progress; similarly, it allows rankings within competition environments like RoboCup. A quantitative assessment of map quality based on an image similarity metric Ψ is introduced here. It is shown through synthetic as well as through real world data that the metric captures intuitive notions of map quality. Furthermore, the metric is compared to a seemingly more straightforward metric based on Least Mean Squared Euclidean distances (LMS-ED) between map points and ground truth. It is shown that both capture intuitive notions of map quality in a similar way, but that Ψ can be computed much more efficiently than the LMS-ED.
2008
Birk, Andreas; Stoyanov, Todor; Nevatia, Yashodhan; Ambrus, Rares; Poppinga, Jann; Pathak, Kaustubh
Terrain Classification for Autonomous Robot Mobility: from Safety, Security Rescue Robotics to Planetary Exploration Proceedings Article
In: Planetary Rovers Workshop, International Conference on Robotics and Automation (ICRA), IEEE, 2008.
@inproceedings{PlanetaryRoverWS-ICRA08,
title = {Terrain Classification for Autonomous Robot Mobility: from Safety, Security Rescue Robotics to Planetary Exploration},
author = {Andreas Birk and Todor Stoyanov and Yashodhan Nevatia and Rares Ambrus and Jann Poppinga and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/374413431_Terrain_Classification_for_Autonomous_Robot_Mobility_from_Safety_Security_Rescue_Robotics_to_Planetary_Exploration},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {Planetary Rovers Workshop, International Conference on Robotics and Automation (ICRA)},
publisher = {IEEE},
abstract = {Mobile robots are increasingly used in unstructured domains without permanent supervision by a human operator. One example is Safety, Security and Rescue Robotics (SSRR) where human operators are a scarce resource. There are additional motivations in this domain to increase robot autonomy, e.g., the desire to decrease the cognitive load on the operator or to allow robot operations when communication to a operator's station fails. Planetary exploration has in this respect much in common with SSRR. Namely, it takes place in unstructured environments and it requires high amounts of autonomy due to the significant delay in communication. Here we present efforts to carry over results from research within SSRR, especially work on terrain classification for autonomous mobility, to planetary exploration. The simple yet efficient approach to terrain classification is based on the Hough transform of planes. The idea is to design a parameter space such that drivable surfaces lead to a strong single response, whereas non-drivable ones lead to data-points spread over the parameter space. The distinction between negotiable and non-negotiable as well as other terrain type is then done by a decision tree. The algorithm is applied in the SSRR domain to 3D data obtained from two different sensors, namely, a near infra-red time of flight camera and a stereo camera. Experimental results are presented for typical indoor as well as outdoor terrains, demonstrating robust realtime detection of drivable ground. The work is then carried over to the planetary exploration domain by using data from the Mars Exploration Rover Mission (MER).},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Mobile robots are increasingly used in unstructured domains without permanent supervision by a human operator. One example is Safety, Security and Rescue Robotics (SSRR) where human operators are a scarce resource. There are additional motivations in this domain to increase robot autonomy, e.g., the desire to decrease the cognitive load on the operator or to allow robot operations when communication to a operator's station fails. Planetary exploration has in this respect much in common with SSRR. Namely, it takes place in unstructured environments and it requires high amounts of autonomy due to the significant delay in communication. Here we present efforts to carry over results from research within SSRR, especially work on terrain classification for autonomous mobility, to planetary exploration. The simple yet efficient approach to terrain classification is based on the Hough transform of planes. The idea is to design a parameter space such that drivable surfaces lead to a strong single response, whereas non-drivable ones lead to data-points spread over the parameter space. The distinction between negotiable and non-negotiable as well as other terrain type is then done by a decision tree. The algorithm is applied in the SSRR domain to 3D data obtained from two different sensors, namely, a near infra-red time of flight camera and a stereo camera. Experimental results are presented for typical indoor as well as outdoor terrains, demonstrating robust realtime detection of drivable ground. The work is then carried over to the planetary exploration domain by using data from the Mars Exploration Rover Mission (MER).
Chonnaparamutt, Winai; Birk, Andreas
A Fuzzy Controller for Autonomous Negotiation of Stairs by a Mobile Robot with Adjustable Tracks Book Chapter
In: Visser, U.; Ribeiro, F.; Ohashi, T.; Dellaert, F. (Ed.): RoboCup 2007: Robot WorldCup XI, Springer, 2008.
@inbook{stairs_rcup07,
title = {A Fuzzy Controller for Autonomous Negotiation of Stairs by a Mobile Robot with Adjustable Tracks},
author = {Winai Chonnaparamutt and Andreas Birk},
editor = {U. Visser and F. Ribeiro and T. Ohashi and F. Dellaert},
url = {https://www.researchgate.net/publication/220797360_A_Fuzzy_Controller_for_Autonomous_Negotiation_of_Stairs_by_a_Mobile_Robot_with_Adjustable_Tracks},
doi = {10.1007/978-3-540-68847-1_17},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {RoboCup 2007: Robot WorldCup XI},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Tracked mobile robots with adjustable support tracks or flippers are popular promising solutions for negotiating rough terrain and 3D obstacles. Though many according robot bases are in principle physically capable of climbing stairs, it is a non-trivial control-task for a remote tele-operator, especially when the user can not directly see the robot like in search and rescue scenarios. To limit training requirements and to ease the cognitive load on operators, respectively to enable fully autonomous rescue robots, we developed a fuzzy controller for this task, which adjusts the drive forces and the posture of the flipper. The design of the controller is guided by observing the strategies of a trained user when tele-operating a robot with unlimited visual information. In doing so, an Open Dynamics Engine (ODE) simulation of our robot is used where the full set of all physical parameters is accessible for analysis. Based on this data, it is shown in several experiments that the controller is not only capable of climbing stairs but that it does so in a more efficient manner than the human user who served as training model.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Tracked mobile robots with adjustable support tracks or flippers are popular promising solutions for negotiating rough terrain and 3D obstacles. Though many according robot bases are in principle physically capable of climbing stairs, it is a non-trivial control-task for a remote tele-operator, especially when the user can not directly see the robot like in search and rescue scenarios. To limit training requirements and to ease the cognitive load on operators, respectively to enable fully autonomous rescue robots, we developed a fuzzy controller for this task, which adjusts the drive forces and the posture of the flipper. The design of the controller is guided by observing the strategies of a trained user when tele-operating a robot with unlimited visual information. In doing so, an Open Dynamics Engine (ODE) simulation of our robot is used where the full set of all physical parameters is accessible for analysis. Based on this data, it is shown in several experiments that the controller is not only capable of climbing stairs but that it does so in a more efficient manner than the human user who served as training model.
Nevatia, Yashodhan; Stoyanov, Todor; Rathnam, Ravi; Pfingsthorn, Max; Markov, Stefan; Ambrus, Rares; Birk, Andreas
Augmented Autonomy: Improving human-robot team performance in Urban Search and Rescue Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2008.
@inproceedings{AugmentedAutonomy-IROS08,
title = {Augmented Autonomy: Improving human-robot team performance in Urban Search and Rescue},
author = {Yashodhan Nevatia and Todor Stoyanov and Ravi Rathnam and Max Pfingsthorn and Stefan Markov and Rares Ambrus and Andreas Birk},
url = {https://www.researchgate.net/publication/224339913_Augmented_Autonomy_Improving_human-robot_team_performance_in_Urban_Search_and_Rescue},
doi = {10.1109/IROS.2008.4651034},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {Exploration of unknown environments remains one of the fundamental problems of mobile robotics. It is also a prime example for a task that can benefit significantly from multi-robot teams. We present an integrated system for semi-autonomous cooperative exploration, augmented by an intuitive user interface for efficient human supervision and control. In this preliminary study we demonstrate the effectiveness of the system as a whole and the intuitive interface in particular. Congruent with previous findings, results confirm that having a human in the loop improves task performance, especially with larger numbers of robots. Specific to our interface, we find that even untrained operators can efficiently manage a decently sized team of robots.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Exploration of unknown environments remains one of the fundamental problems of mobile robotics. It is also a prime example for a task that can benefit significantly from multi-robot teams. We present an integrated system for semi-autonomous cooperative exploration, augmented by an intuitive user interface for efficient human supervision and control. In this preliminary study we demonstrate the effectiveness of the system as a whole and the intuitive interface in particular. Congruent with previous findings, results confirm that having a human in the loop improves task performance, especially with larger numbers of robots. Specific to our interface, we find that even untrained operators can efficiently manage a decently sized team of robots.
Pathak, Kaustubh; Birk, Andreas; Poppinga, Jann
Subpixel Depth Accuracy with a Time of Flight Sensor using Multimodal Gaussian Analysis Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2008.
@inproceedings{SubPixelDepth-IROS08,
title = {Subpixel Depth Accuracy with a Time of Flight Sensor using Multimodal Gaussian Analysis},
author = {Kaustubh Pathak and Andreas Birk and Jann Poppinga},
url = {https://www.researchgate.net/publication/224339728_Sub-pixel_Depth_Accuracy_with_a_Time_of_Flight_Sensor_using_Multimodal_Gaussian_Analysis},
doi = {10.1109/IROS.2008.4650841},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {Pixel array based 3D range sensors like time of flight sensors (e.g. Swiss-ranger) are commonly used for spatial mapping. Analogous to a laser range finder beam, each pixel measures distances within its conical field of view. However, unlike laser range finders, these sensors use incoherent visible or near visible light for range measurement. This, combined with their relatively long maximum ranges, means that the convenient thin beam assumption can no longer be used to update an occupancy grid map using a probabilistic sensor model. We present an analysis of a scenario where a pixel beam is intersected by more than one object and where a unimodal probabilistic distribution assumption causes spurious objects to appear in the detected scene. This adversely affects path-planning based on such maps because perfectly good escape routes are blocked off. A two step multimodal Gaussian mixture model based procedure is presented which is able to detect multiple obstacles per pixel and hence ameliorate the problem. The results of experiments done using a time of flight based sensor are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Pixel array based 3D range sensors like time of flight sensors (e.g. Swiss-ranger) are commonly used for spatial mapping. Analogous to a laser range finder beam, each pixel measures distances within its conical field of view. However, unlike laser range finders, these sensors use incoherent visible or near visible light for range measurement. This, combined with their relatively long maximum ranges, means that the convenient thin beam assumption can no longer be used to update an occupancy grid map using a probabilistic sensor model. We present an analysis of a scenario where a pixel beam is intersected by more than one object and where a unimodal probabilistic distribution assumption causes spurious objects to appear in the detected scene. This adversely affects path-planning based on such maps because perfectly good escape routes are blocked off. A two step multimodal Gaussian mixture model based procedure is presented which is able to detect multiple obstacles per pixel and hence ameliorate the problem. The results of experiments done using a time of flight based sensor are presented.
Pfingsthorn, Max; Birk, Andreas
Efficiently Communicating Map Updates with the Pose Graph Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), IEEE Press, 2008.
@inproceedings{CommPoseGraph-IROS08,
title = {Efficiently Communicating Map Updates with the Pose Graph},
author = {Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/224340060_Efficiently_Communicating_Map_Updates_with_the_Pose_Graph},
doi = {10.1109/IROS.2008.4651182},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
publisher = {IEEE Press},
abstract = {Robot mapping is a task that can benefit a lot from cooperative multi-robot systems. In multi-robot simultaneous localization and mapping (SLAM), it becomes very important how efficiently a given map can be shared among the robot team. To this end, the recently proposed pose graph map representation is used, adapted for use in a particle filter based mapping algorithm, and compared to the standard occupancy grid representation. Through analysis of corner cases as well as experiments with real robot data, the two map representations are thoroughly compared. It is shown that the pose graph representation allows for much more efficient communication of map updates than the standard occupancy grid.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Robot mapping is a task that can benefit a lot from cooperative multi-robot systems. In multi-robot simultaneous localization and mapping (SLAM), it becomes very important how efficiently a given map can be shared among the robot team. To this end, the recently proposed pose graph map representation is used, adapted for use in a particle filter based mapping algorithm, and compared to the standard occupancy grid representation. Through analysis of corner cases as well as experiments with real robot data, the two map representations are thoroughly compared. It is shown that the pose graph representation allows for much more efficient communication of map updates than the standard occupancy grid.
Poppinga, Jann; Birk, Andreas; Pathak, Kaustubh
Hough based Terrain Classification for Realtime Detection of Drivable Ground Journal Article
In: Journal of Field Robotics, vol. 25, no. 1-2, pp. 67-88, 2008.
@article{HoughTerrainClassification-JFR07,
title = {Hough based Terrain Classification for Realtime Detection of Drivable Ground},
author = {Jann Poppinga and Andreas Birk and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/220647997_Hough_based_terrain_classification_for_realtime_detection_of_drivable_ground},
doi = {10.1002/rob.20227},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
journal = {Journal of Field Robotics},
volume = {25},
number = {1-2},
pages = {67-88},
abstract = {The usability of mobile robots for surveillance, search and rescue missions can be significantly improved by intelligent functionalities decreasing the cognitive loadon the operator or even allowing autonomous operations, e.g., when communication fails. Mobility in this regard is not only a mechatronic problem but also a perception, modeling and planning challenge. Here, the perception issue of detecting drivable ground is addressed, an important issue for safety, security, and rescue robots, which have to operate in a vast range of unstructured, challenging environments. The simple yet efficient approach is based on the Hough transform of planes. The idea is to design the parameter space such that drivable surfaces can be easily detected by the number of hits in the bins corresponding to drivability. A decision tree on the bin properties increases robustness as it allows to handle uncertainties, especially sensor noise. In addition to the binary distinction of drivable/non-drivable ground, a classification of terrain types is possible. The algorithm is applied to 3D data obtained from two different sensors, namely, a time-of-flight camera and a stereo camera. Experimental results are presented for indoor and outdoor terrains, demonstrating robust realtime detection of drivable ground. Seven datasets recorded under very varying conditions are used. About 6,800 snapshots of range data are processed in total. It is shown that drivability can be robustly detected with success rates ranging between 83% and 100%. Computation is extremely fast in the order of 5 to 50 msec.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The usability of mobile robots for surveillance, search and rescue missions can be significantly improved by intelligent functionalities decreasing the cognitive loadon the operator or even allowing autonomous operations, e.g., when communication fails. Mobility in this regard is not only a mechatronic problem but also a perception, modeling and planning challenge. Here, the perception issue of detecting drivable ground is addressed, an important issue for safety, security, and rescue robots, which have to operate in a vast range of unstructured, challenging environments. The simple yet efficient approach is based on the Hough transform of planes. The idea is to design the parameter space such that drivable surfaces can be easily detected by the number of hits in the bins corresponding to drivability. A decision tree on the bin properties increases robustness as it allows to handle uncertainties, especially sensor noise. In addition to the binary distinction of drivable/non-drivable ground, a classification of terrain types is possible. The algorithm is applied to 3D data obtained from two different sensors, namely, a time-of-flight camera and a stereo camera. Experimental results are presented for indoor and outdoor terrains, demonstrating robust realtime detection of drivable ground. Seven datasets recorded under very varying conditions are used. About 6,800 snapshots of range data are processed in total. It is shown that drivability can be robustly detected with success rates ranging between 83% and 100%. Computation is extremely fast in the order of 5 to 50 msec.
Poppinga, Jann; Vaskevicius, Narunas; Birk, Andreas; Pathak, Kaustubh
Fast Plane Detection and Polygonalization in noisy 3D Range Images Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), pp. 3378 - 3383, IEEE Press, 2008.
@inproceedings{PlaneDetect-IROS08,
title = {Fast Plane Detection and Polygonalization in noisy 3D Range Images},
author = {Jann Poppinga and Narunas Vaskevicius and Andreas Birk and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/224339625_Fast_Plane_Detection_and_Polygonalization_in_Noisy_3D_Range_Images},
doi = {10.1109/IROS.2008.4650729},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
pages = {3378 - 3383},
publisher = {IEEE Press},
abstract = {A fast but nevertheless accurate approach for surface extraction from noisy 3D point clouds is presented. It consists of two parts, namely a plane fitting and a polygonalization step. Both exploit the sequential nature of 3D data acquisition on mobile robots in form of range images. For the plane fitting, this is used to revise the standard mathematical formulation to an incremental version, which allows a linear computation. For the polygonalization, the neighborhood relation in range images is exploited. Experiments are presented using a time-of-flight range camera in form of a Swissranger SR-3000. Results include lab scenes as well as data from two runs of the rescue robot league at the RoboCup German Open 2007 with 1,414, respectively 2,343 sensor snapshots. The 36*10^6, respectively 59*10^6 points from the two point clouds are reduced to about 14*10^3, respectively 23*10^3 planes with only about 0.2 sec of total computation time per snapshot while the robot moves along. Uncertainty analysis of the computed plane parameters is presented as well.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A fast but nevertheless accurate approach for surface extraction from noisy 3D point clouds is presented. It consists of two parts, namely a plane fitting and a polygonalization step. Both exploit the sequential nature of 3D data acquisition on mobile robots in form of range images. For the plane fitting, this is used to revise the standard mathematical formulation to an incremental version, which allows a linear computation. For the polygonalization, the neighborhood relation in range images is exploited. Experiments are presented using a time-of-flight range camera in form of a Swissranger SR-3000. Results include lab scenes as well as data from two runs of the rescue robot league at the RoboCup German Open 2007 with 1,414, respectively 2,343 sensor snapshots. The 36*10^6, respectively 59*10^6 points from the two point clouds are reduced to about 14*10^3, respectively 23*10^3 planes with only about 0.2 sec of total computation time per snapshot while the robot moves along. Uncertainty analysis of the computed plane parameters is presented as well.
Schwertfeger, Soeren; Poppinga, Jann; Birk, Andreas
Towards Object Classification using 3D Sensor Data Proceedings Article
In: ECSIS Symposium on Learning and Adaptive Behaviors for Robotic Systems (LAB-RS), IEEE, 2008.
@inproceedings{ObjectClassification-LABRS08,
title = {Towards Object Classification using 3D Sensor Data},
author = {Soeren Schwertfeger and Jann Poppinga and Andreas Birk},
url = {https://www.researchgate.net/publication/4365336_Towards_Object_Classification_Using_3D_Sensor_Data},
doi = {10.1109/LAB-RS.2008.28},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {ECSIS Symposium on Learning and Adaptive Behaviors for Robotic Systems (LAB-RS)},
publisher = {IEEE},
abstract = {This paper presents an approach to classify objects using 3D sensor data and an evolutionary algorithm. An important by-product of this classification is, that additionally certain properties and the pose in space of this object are determined. The reproductive perception paradigm is used utilizing an evolutionary strategy. Two sub-approaches are discussed using different representations of the 3D data. The first one uses depth images while the second one uses point clouds stored in a special octree. The approaches will be demonstrated in experiments with simulated and real data.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents an approach to classify objects using 3D sensor data and an evolutionary algorithm. An important by-product of this classification is, that additionally certain properties and the pose in space of this object are determined. The reproductive perception paradigm is used utilizing an evolutionary strategy. Two sub-approaches are discussed using different representations of the 3D data. The first one uses depth images while the second one uses point clouds stored in a special octree. The approaches will be demonstrated in experiments with simulated and real data.
Varsadan, Ioana; Birk, Andreas; Pfingsthorn, Max; Schwertfeger, Sören; Pathak, Kaustubh
The Jacobs map analysis toolkit Proceedings Article
In: Workshop on Experimental Methodology and Benchmarking in Robotics Research, Robotics Science and Systems (RSS), 2008.
@inproceedings{JacobsMapAnalysisToolkit-EMBRR08,
title = {The Jacobs map analysis toolkit},
author = {Ioana Varsadan and Andreas Birk and Max Pfingsthorn and Sören Schwertfeger and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/263258343_The_Jacobs_map_analysis_toolkit},
year = {2008},
date = {2008-01-01},
urldate = {2008-01-01},
booktitle = {Workshop on Experimental Methodology and Benchmarking in Robotics Research, Robotics Science and Systems (RSS)},
abstract = {A quantitative assessment of the quality of a robot generated map is of high interest for many reasons. First of all, it allows individual researchers to quantify the quality of their mapping approach and to study the effects of system specific choices like different parameter values in an objective way. Second, it allows peer groups to rank the quality of their different approaches to determine scientific progress; similarly, it allows rankings within competition environments like RoboCup. The Jacobs Map Analysis Toolkit is presented here, which is open source and freely available. It allows the visual as well as the numerical comparison of maps. The toolkit also includes a ground truth designer, which facilitates the generation of reference data for environment based on standard NIST test elements. One quantitative assessment of map quality in the toolkit is for example based on the image similarity metric $Psi$. It is shown through synthetic as well as through real world data that the metric captures intuitive notions of map quality. Furthermore, the metric is compared to a seemingly more straightforward metric based on Least Mean Squared Euclidean distances (LMS-ED) between map points and ground truth. It is shown that both capture intuitive notions of map quality in a similar way, but that $Psi$ can be computed much more efficiently than the LMS-ED.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
A quantitative assessment of the quality of a robot generated map is of high interest for many reasons. First of all, it allows individual researchers to quantify the quality of their mapping approach and to study the effects of system specific choices like different parameter values in an objective way. Second, it allows peer groups to rank the quality of their different approaches to determine scientific progress; similarly, it allows rankings within competition environments like RoboCup. The Jacobs Map Analysis Toolkit is presented here, which is open source and freely available. It allows the visual as well as the numerical comparison of maps. The toolkit also includes a ground truth designer, which facilitates the generation of reference data for environment based on standard NIST test elements. One quantitative assessment of map quality in the toolkit is for example based on the image similarity metric $Psi$. It is shown through synthetic as well as through real world data that the metric captures intuitive notions of map quality. Furthermore, the metric is compared to a seemingly more straightforward metric based on Least Mean Squared Euclidean distances (LMS-ED) between map points and ground truth. It is shown that both capture intuitive notions of map quality in a similar way, but that $Psi$ can be computed much more efficiently than the LMS-ED.
2007
Albu, Diana; Birk, Andreas; Dobrev, Petar; Gammoh, Farah; Giurgiu, Andrei; Mihut, Sergiu-Cristian; Minzu, Bogdan; Pascanu, Razvan; Schwertfeger, Soeren; Stan, Alexandru; Videv, Stefan
Fast Prototyping of an Autonomous Underwater Vehicle (AUV) with the CubeSystem Proceedings Article
In: 6th International Symposium on Intelligent Autonomous Vehicles (IAV 2007), IFAC, 2007.
@inproceedings{IUB-Atlas-AUV-IAV07,
title = {Fast Prototyping of an Autonomous Underwater Vehicle (AUV) with the CubeSystem},
author = {Diana Albu and Andreas Birk and Petar Dobrev and Farah Gammoh and Andrei Giurgiu and Sergiu-Cristian Mihut and Bogdan Minzu and Razvan Pascanu and Soeren Schwertfeger and Alexandru Stan and Stefan Videv},
url = {https://www.researchgate.net/publication/302072251_FAST_PROTOTYPING_OF_AN_AUTONOMOUS_UNDERWATER_VEHICLE_WITH_THE_CUBESYSTEM},
doi = {10.3182/20070903-3-FR-2921.00016},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {6th International Symposium on Intelligent Autonomous Vehicles (IAV 2007)},
publisher = {IFAC},
abstract = {This paper describes the Jacobs autonomous underwater vehicle (Jacobs-Atlas-AUV). The vehicle has been largely developed using the CubeSystem, a collection of hard- and software components for fast robot prototyping. The CubeSystem has previously been used for the development of various mobile robots, but never for an underwater system. As the development of the Jacobs-Atlas-AUV shows, a significant amount of component reuse across significantly different application domains is possible for autonomous robots.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the Jacobs autonomous underwater vehicle (Jacobs-Atlas-AUV). The vehicle has been largely developed using the CubeSystem, a collection of hard- and software components for fast robot prototyping. The CubeSystem has previously been used for the development of various mobile robots, but never for an underwater system. As the development of the Jacobs-Atlas-AUV shows, a significant amount of component reuse across significantly different application domains is possible for autonomous robots.
Birk, Andreas
Spatial Knowledge Processing within the Reproductive Perception Paradigm Proceedings Article
In: Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems, AAAI Spring Symposium, Technical Report SS-07-01, Stanford, California, USA, AAAI, 2007.
@inproceedings{birk-SS07-repropercep,
title = {Spatial Knowledge Processing within the Reproductive Perception Paradigm},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/221251220_Spatial_Knowledge_Processing_within_the_Reproductive_Perception_Paradigm},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {Control Mechanisms for Spatial Knowledge Processing in Cognitive / Intelligent Systems, AAAI Spring Symposium, Technical Report SS-07-01, Stanford, California, USA},
publisher = {AAAI},
series = {Spring Symposium},
abstract = {Reproductive perception is a novel approach to perception. It is based on the assumption that perception is predominantly a generative process, i.e., that models that represent hypotheses about the current state of the environment generate so-called pseudo-sensor data, which is matched against the actual sensor data to determine and improve the correctness of these hypotheses. This is in contrast to the view that perception is mainly a reductive process where large amounts of sensor data are processed until a compact representation is achieved. Several successful examples of using this approach for spatial world-modeling are presented here. The first one deals with a robot arm and a camera to learn eye hand coordination. In the second example, the successful detection and 3D localization of humans in single 2D thermal images is described. In the last but not least example, work in progress on the generation and usage of compact 3D models of unstructured environments on a mobile robot is presented. The reproductive perception approach in all three examples does not only influence the way the spatial knowledge is generated and represented, but also its usage especially with respect to the classification and recognition of objects.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Reproductive perception is a novel approach to perception. It is based on the assumption that perception is predominantly a generative process, i.e., that models that represent hypotheses about the current state of the environment generate so-called pseudo-sensor data, which is matched against the actual sensor data to determine and improve the correctness of these hypotheses. This is in contrast to the view that perception is mainly a reductive process where large amounts of sensor data are processed until a compact representation is achieved. Several successful examples of using this approach for spatial world-modeling are presented here. The first one deals with a robot arm and a camera to learn eye hand coordination. In the second example, the successful detection and 3D localization of humans in single 2D thermal images is described. In the last but not least example, work in progress on the generation and usage of compact 3D models of unstructured environments on a mobile robot is presented. The reproductive perception approach in all three examples does not only influence the way the spatial knowledge is generated and represented, but also its usage especially with respect to the classification and recognition of objects.
Birk, Andreas; Pathak, Kausthub; Chonnaparamutt, Winai; Schwertfeger, Soeren; Delchev, Ivan; Markov, Stefan; Rathnam, Ravi
The IUB 2006 Rescue Robot Team Book Chapter
In: Lakemeyer, G.; Sklar, E.; Sorrenti, D.; Takahashi, T. (Ed.): RoboCup 2006: Robot Soccer World Cup X, vol. 4434, Springer, 2007.
@inbook{rescueteam_rcup06,
title = {The IUB 2006 Rescue Robot Team},
author = {Andreas Birk and Kausthub Pathak and Winai Chonnaparamutt and Soeren Schwertfeger and Ivan Delchev and Stefan Markov and Ravi Rathnam},
editor = {G. Lakemeyer and E. Sklar and D. Sorrenti and T. Takahashi},
url = {https://www.researchgate.net/publication/260291904_The_IUB_2006_Rescue_Robot_Team},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {RoboCup 2006: Robot Soccer World Cup X},
volume = {4434},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas; Pathak, Kaustubh; Poppinga, Jann; Schwertfeger, Soeren; Chonnaparamutt, Winai
Intelligent Behaviors in Outdoor Environments Proceedings Article
In: 13th International Conference on Robotics and Applications, Special Session on Outdoor Robotics - Taking Robots off road, IASTED, 2007.
@inproceedings{BirkOutdoorRobotics07,
title = {Intelligent Behaviors in Outdoor Environments},
author = {Andreas Birk and Kaustubh Pathak and Jann Poppinga and Soeren Schwertfeger and Winai Chonnaparamutt},
url = {https://www.researchgate.net/publication/234815332_Intelligent_behaviors_in_outdoor_environments},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {13th International Conference on Robotics and Applications, Special Session on Outdoor Robotics - Taking Robots off road},
publisher = {IASTED},
abstract = {The Jacobs Rugbots - derived from "rugged robot" - are capable of intelligent behaviors up to full autonomy in outdoor environments. One crucial aspect for the according intelligent behaviors is the capability to detect negotiable terrain in a very fast and robust manner. Our approach uses range data from 3D sensors like a time-of-flight Swissranger, respectively a stereo camera. The range data is processed by a Hough transform with a special 3D parameter space, which is chosen such that its bins correspond to planes that can be negotiated by the robot. Extensive experiments with different prototypical indoor and outdoor ground types are presented, which can be clearly classified with respect to drivability. Seven datasets, two recorded indoor and five outdoor under very varying conditions, are used for testing. Drivability is robustly detected with success rates ranging between 83% and 100% for the Swissranger and between 98% and 100% for the stereo camera. The complete processing time per snapshot is in the order of 5 to 50 msec.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The Jacobs Rugbots - derived from "rugged robot" - are capable of intelligent behaviors up to full autonomy in outdoor environments. One crucial aspect for the according intelligent behaviors is the capability to detect negotiable terrain in a very fast and robust manner. Our approach uses range data from 3D sensors like a time-of-flight Swissranger, respectively a stereo camera. The range data is processed by a Hough transform with a special 3D parameter space, which is chosen such that its bins correspond to planes that can be negotiated by the robot. Extensive experiments with different prototypical indoor and outdoor ground types are presented, which can be clearly classified with respect to drivability. Seven datasets, two recorded indoor and five outdoor under very varying conditions, are used for testing. Drivability is robustly detected with success rates ranging between 83% and 100% for the Swissranger and between 98% and 100% for the stereo camera. The complete processing time per snapshot is in the order of 5 to 50 msec.
Birk, Andreas; Pathak, Kaustubh; Poppinga, Jann; Schwertfeger, Sören; Pfingsthorn, Max; Bülow, Heiko
The Jacobs Test Arena for Safety, Security, and Rescue Robotics (SSRR) Proceedings Article
In: WS on Performance Evaluation and Benchmarking for Intelligent Robots and Systems, Intern. Conf. on Intelligent Robots and Systems (IROS), 2007.
@inproceedings{JacobsArenaIROS07,
title = {The Jacobs Test Arena for Safety, Security, and Rescue Robotics (SSRR)},
author = {Andreas Birk and Kaustubh Pathak and Jann Poppinga and Sören Schwertfeger and Max Pfingsthorn and Heiko Bülow},
url = {https://www.researchgate.net/publication/263258287_The_Jacobs_Test_Arena_for_Safety_Security_and_Rescue_Robotics_SSRR},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {WS on Performance Evaluation and Benchmarking for Intelligent Robots and Systems, Intern. Conf. on Intelligent Robots and Systems (IROS)},
abstract = {The Jacobs University Bremen features a special test site for mobile robots. It consists of two large arenas with test elements, which are particularly suited to evaluate the performance of systems operating in challenging domains like Security, Safety, and Rescue Robotics (SSRR). The site is one of only six worldwide, which has been established in close cooperation with the US National Institute of Standards and Technology (NIST). This paper presents the Jacobs arenas and a set of metrics for evaluating the mobility, sensing capability, and onboard intelligence of robots. The tests are illustrated by using a Jacobs rescue robot, which is equipped with state of the art sensors. The relative strengths and weaknesses of these sensors are evaluated in a variety of situations; many of them are typically encountered in SSRR applications.
},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The Jacobs University Bremen features a special test site for mobile robots. It consists of two large arenas with test elements, which are particularly suited to evaluate the performance of systems operating in challenging domains like Security, Safety, and Rescue Robotics (SSRR). The site is one of only six worldwide, which has been established in close cooperation with the US National Institute of Standards and Technology (NIST). This paper presents the Jacobs arenas and a set of metrics for evaluating the mobility, sensing capability, and onboard intelligence of robots. The tests are illustrated by using a Jacobs rescue robot, which is equipped with state of the art sensors. The relative strengths and weaknesses of these sensors are evaluated in a variety of situations; many of them are typically encountered in SSRR applications.
Chonnaparamutt, Winai; Birk, Andreas
A New Mechatronic Component for Adjusting the Footprint of Tracked Rescue Robots Book Chapter
In: Lakemeyer, G.; Sklar, E.; Sorrenti, D.; Takahashi, T. (Ed.): RoboCup 2006: Robot WorldCup X, vol. 4434, Springer, 2007.
@inbook{flipper_rcup06,
title = {A New Mechatronic Component for Adjusting the Footprint of Tracked Rescue Robots},
author = {Winai Chonnaparamutt and Andreas Birk},
editor = {G. Lakemeyer and E. Sklar and D. Sorrenti and T. Takahashi},
url = {https://www.researchgate.net/publication/220797193_A_New_Mechatronic_Component_for_Adjusting_the_Footprint_of_Tracked_Rescue_Robots},
doi = {10.1007/978-3-540-74024-7_45},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {RoboCup 2006: Robot WorldCup X},
volume = {4434},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {There is no ideal footprint for a rescue robot. In some situations, for example when climbing up a rubble pile or stairs, the footprint has to be large to maximize traction and to prevent tilting over. In other situations, for example when negotiating narrow passages or doorways, the footprint has to be small to prevent to get stuck. The common approach is to use flippers, i.e., additional support tracks that can change their posture relative to the main locomotion tracks. Here a novel mechatronic design for flippers is presented that overcomes a significant drawback in the state of the art approaches, namely the large forces in the joint between main locomotion tracks and flippers. Instead of directly driving this joint to change the posture, a link mechanism driven by a ballscrew is used. In this paper, a formal analysis of the new mechanism is presented including a comparison to the state of the art. Furthermore, a concrete implementation and results from practical experiments that support the formal analysis are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
There is no ideal footprint for a rescue robot. In some situations, for example when climbing up a rubble pile or stairs, the footprint has to be large to maximize traction and to prevent tilting over. In other situations, for example when negotiating narrow passages or doorways, the footprint has to be small to prevent to get stuck. The common approach is to use flippers, i.e., additional support tracks that can change their posture relative to the main locomotion tracks. Here a novel mechatronic design for flippers is presented that overcomes a significant drawback in the state of the art approaches, namely the large forces in the joint between main locomotion tracks and flippers. Instead of directly driving this joint to change the posture, a link mechanism driven by a ballscrew is used. In this paper, a formal analysis of the new mechanism is presented including a comparison to the state of the art. Furthermore, a concrete implementation and results from practical experiments that support the formal analysis are presented.
Delchev, Ivan; Birk, Andreas
Vectorization of Grid Maps by an Evolutionary Algorithm Book Chapter
In: Lakemeyer, G.; Sklar, E.; Sorrenti, D.; Takahashi, T. (Ed.): RoboCup 2006: Robot WorldCup X, vol. 4434, Springer, 2007.
@inbook{mapvectorization_rcup06,
title = {Vectorization of Grid Maps by an Evolutionary Algorithm},
author = {Ivan Delchev and Andreas Birk},
editor = {G. Lakemeyer and E. Sklar and D. Sorrenti and T. Takahashi},
url = {https://www.researchgate.net/publication/220797697_Vectorization_of_Grid_Maps_by_an_Evolutionary_Algorithm},
doi = {10.1007/978-3-540-74024-7_46},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {RoboCup 2006: Robot WorldCup X},
volume = {4434},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Mapping is a fundamental topic for robotics in general and in particular for rescue robotics where the provision of information about the location of victims is a core task. Occupancy grids are the standard way of generating and representing maps, i.e., in form of raster data. But vector representations are for many reasons, especially due to their compactness and the possibility to use very efficient computational geometry algorithms, highly desirable for many applications. Here a novel method for vectorization is presented that is intended to work particularly well with maps. It is based on an evolutionary algorithm that generates vector code for a so to say drawing program. The output of the evolving vector code is compared to the input grid map via a special similarity function as fitness. Experiments are presented that indicate that the approach is indeed a successful method to extract vector data out of grid maps.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Mapping is a fundamental topic for robotics in general and in particular for rescue robotics where the provision of information about the location of victims is a core task. Occupancy grids are the standard way of generating and representing maps, i.e., in form of raster data. But vector representations are for many reasons, especially due to their compactness and the possibility to use very efficient computational geometry algorithms, highly desirable for many applications. Here a novel method for vectorization is presented that is intended to work particularly well with maps. It is based on an evolutionary algorithm that generates vector code for a so to say drawing program. The output of the evolving vector code is compared to the input grid map via a special similarity function as fitness. Experiments are presented that indicate that the approach is indeed a successful method to extract vector data out of grid maps.
Markov, Stefan; Birk, Andreas
Detecting Humans in 2D Thermal Images by Generating 3D Models Proceedings Article
In: Hertzberg, Joachim; Beetz, Michael; Englert, Roman (Ed.): KI 2007: Advances in Artificial Intelligence, Springer, 2007.
@inproceedings{HumanRec-KI07,
title = {Detecting Humans in 2D Thermal Images by Generating 3D Models},
author = {Stefan Markov and Andreas Birk},
editor = {Joachim Hertzberg and Michael Beetz and Roman Englert},
url = {https://www.researchgate.net/publication/221562427_Detecting_Humans_in_2D_Thermal_Images_by_Generating_3D_Models},
doi = {10.1007/978-3-540-74565-5_23},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {KI 2007: Advances in Artificial Intelligence},
volume = {4667},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {There are two significant challenges to standard approaches to detect humans through computer vision. First, scenarios when the poses and postures of the humans are completely unpredictable. Second, situations when there are many occlusions, i.e., only parts of the body are visible. Here a novel approach to perception is presented where a complete 3D scene model is learned on the fly to represent a 2D snapshot. In doing so, an evolutionary algorithm generates pieces of 3D code that are rendered and the resulting images are compared to the current camera picture via an image similarity function. Based on the feedback of this fitness function, a crude but very fast online evolution generates an approximate D model of the environment where non-human objects are represented by boxes. The key point is that 3D models of humans are available as code sniplets to the EA, which can use them to represent human shapes or portions of them if they are in the image. Results from experiments with real world data from a search and rescue application using a thermal camera are presented.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
There are two significant challenges to standard approaches to detect humans through computer vision. First, scenarios when the poses and postures of the humans are completely unpredictable. Second, situations when there are many occlusions, i.e., only parts of the body are visible. Here a novel approach to perception is presented where a complete 3D scene model is learned on the fly to represent a 2D snapshot. In doing so, an evolutionary algorithm generates pieces of 3D code that are rendered and the resulting images are compared to the current camera picture via an image similarity function. Based on the feedback of this fitness function, a crude but very fast online evolution generates an approximate D model of the environment where non-human objects are represented by boxes. The key point is that 3D models of humans are available as code sniplets to the EA, which can use them to represent human shapes or portions of them if they are in the image. Results from experiments with real world data from a search and rescue application using a thermal camera are presented.
Pathak, Kaustubh; Birk, Andreas; Schwertfeger, Soeren; Delchev, Ivan; Markov, Stefan
Fully Autonomous Operations of a Jacobs Rugbot in the RoboCup Rescue Robot League 2006 Proceedings Article
In: International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2007.
@inproceedings{SSRR07-JacobsAutonomy,
title = {Fully Autonomous Operations of a Jacobs Rugbot in the RoboCup Rescue Robot League 2006},
author = {Kaustubh Pathak and Andreas Birk and Soeren Schwertfeger and Ivan Delchev and Stefan Markov},
url = {https://www.researchgate.net/publication/4290684_Fully_Autonomous_Operations_of_a_Jacobs_Rugbot_in_the_RoboCup_Rescue_Robot_League_2006},
doi = {10.1109/SSRR.2007.4381267},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {This paper describes the latest version of an integrated hardware and software framework developed for autonomous operation of rescue robots. The successful operation of an autonomous rugbot -short for "rugged robot -was especially demonstrated during several runs at the RoboCup world championship 2006 in Bremen. The design of the autonomous system is described in detail with an emphasis on extendibility and the specific requirements of a typical unstructured rescue scenario.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper describes the latest version of an integrated hardware and software framework developed for autonomous operation of rescue robots. The successful operation of an autonomous rugbot -short for "rugged robot -was especially demonstrated during several runs at the RoboCup world championship 2006 in Bremen. The design of the autonomous system is described in detail with an emphasis on extendibility and the specific requirements of a typical unstructured rescue scenario.
Pathak, Kaustubh; Birk, Andreas; Schwertfeger, Soeren; Poppinga, Jann
3D Forward Sensor Modeling and Application to Occupancy Grid Based Sensor Fusion Proceedings Article
In: International Conference on Intelligent Robots and Systems (IROS), pp. 2059 - 2064, IEEE Press, 2007.
@inproceedings{ForwardModelFusion-IROS07,
title = {3D Forward Sensor Modeling and Application to Occupancy Grid Based Sensor Fusion},
author = {Kaustubh Pathak and Andreas Birk and Soeren Schwertfeger and Jann Poppinga},
url = {https://www.researchgate.net/publication/224296646_3D_forward_sensor_modeling_and_application_to_occupancy_grid_based_sensor_fusion},
doi = {10.1109/IROS.2007.4399406},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {International Conference on Intelligent Robots and Systems (IROS)},
pages = {2059 - 2064},
publisher = {IEEE Press},
abstract = {This paper presents a new technique for the update of a probabilistic spatial occupancy grid map using a forward sensor model. Unlike currently popular inverse sensor models, forward sensor models can be found experimentally and can represent sensor characteristics better. The formulation is applicable to both 2D and 3D range sensors and does not have some of the theoretical and practical problems associated with the current approaches which use forward models. As an illustration of this procedure, a new prototype 3D forward sensor model is derived using a beam represented as a spherical sector. Furthermore, this model is used for fusion of point-clouds obtained from different 3D sensors, in particular, time-of-flight sensors (Swiss-ranger, laser range finders), and stereo vision cameras. Several techniques are described for an efficient data-structure representation and implementation. The range beams from different sensors are fused in a common local Cartesian occupancy map. Experimental results of this fusion are presented and evaluated using Hough-transform performed on the grid.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
This paper presents a new technique for the update of a probabilistic spatial occupancy grid map using a forward sensor model. Unlike currently popular inverse sensor models, forward sensor models can be found experimentally and can represent sensor characteristics better. The formulation is applicable to both 2D and 3D range sensors and does not have some of the theoretical and practical problems associated with the current approaches which use forward models. As an illustration of this procedure, a new prototype 3D forward sensor model is derived using a beam represented as a spherical sector. Furthermore, this model is used for fusion of point-clouds obtained from different 3D sensors, in particular, time-of-flight sensors (Swiss-ranger, laser range finders), and stereo vision cameras. Several techniques are described for an efficient data-structure representation and implementation. The range beams from different sensors are fused in a common local Cartesian occupancy map. Experimental results of this fusion are presented and evaluated using Hough-transform performed on the grid.
Poppinga, Jann; Pfingsthorn, Max; Schwertfeger, Soeren; Pathak, Kaustubh; Birk, Andreas
Optimized Octtree Datastructure and Access Methods for 3D Mapping Proceedings Article
In: IEEE Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2007.
@inproceedings{SSRR07-3Ddatastructures,
title = {Optimized Octtree Datastructure and Access Methods for 3D Mapping},
author = {Jann Poppinga and Max Pfingsthorn and Soeren Schwertfeger and Kaustubh Pathak and Andreas Birk},
url = {https://www.researchgate.net/publication/4290690_Optimized_Octtree_Datastructure_and_Access_Methods_for_3D_Mapping},
doi = {10.1109/SSRR.2007.4381275},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {IEEE Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {3D maps are becoming increasingly important for robots operating in real world scenarios. As occupancy grids are a popular standard in 2D mapping, it is a natural choice to extend them to spatial representation in 3D. But this approach suffers from large memory requirements that render 3D grids unfeasible for realistic mapping applications. A well-known alternative is quadtrees in 2D, respectively octrees for 3D, where cells that contain identical information are collapsed into a single node in a tree. But octrees only allow relative slow access to the spatial information stored in them. Here, a very efficient implementation of an octree is presented. It is based on technical optimizations as well as newly introduced heuristics like the exploitation of typical access patterns. It is shown that the optimizations are indeed beneficial through experiments with real world 3D sensor data. In addition to basic access methods, a very fast nearest neighbor operation - as needed for example for registration in SLAM or map merging algorithms - for octrees is presented. It is shown in experiments with real world data that this nearest neighbor operation outperforms a comparable operation on occupancy grids by far.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
3D maps are becoming increasingly important for robots operating in real world scenarios. As occupancy grids are a popular standard in 2D mapping, it is a natural choice to extend them to spatial representation in 3D. But this approach suffers from large memory requirements that render 3D grids unfeasible for realistic mapping applications. A well-known alternative is quadtrees in 2D, respectively octrees for 3D, where cells that contain identical information are collapsed into a single node in a tree. But octrees only allow relative slow access to the spatial information stored in them. Here, a very efficient implementation of an octree is presented. It is based on technical optimizations as well as newly introduced heuristics like the exploitation of typical access patterns. It is shown that the optimizations are indeed beneficial through experiments with real world 3D sensor data. In addition to basic access methods, a very fast nearest neighbor operation - as needed for example for registration in SLAM or map merging algorithms - for octrees is presented. It is shown in experiments with real world data that this nearest neighbor operation outperforms a comparable operation on occupancy grids by far.
Rooker, Martijn N.; Birk, Andreas
Multi-robot exploration under the constraints of wireless networking Journal Article
In: Control Engineering Practice, vol. 15, no. 4, pp. 435-445, 2007.
@article{CommExpore_CEP06,
title = {Multi-robot exploration under the constraints of wireless networking},
author = {Martijn N. Rooker and Andreas Birk},
url = {https://www.researchgate.net/publication/222818057_Multi-robot_exploration_under_the_constraints_of_wireless_networking},
doi = {10.1016/j.conengprac.2006.08.007},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
journal = {Control Engineering Practice},
volume = {15},
number = {4},
pages = {435-445},
abstract = {A multi-robot system can be highly beneficial for exploration, which is a core robotics task. Application domains include, for example, surveillance, reconnaissance, planetary exploration or rescue missions. When using a team of robots, the overall performance can be much faster and more robust. In this article, an approach to multi-robot exploration is presented that takes the constraints of wireless networking into account. An algorithm is introduced based on a population that samples the possible moves of all robots and a utility to select the best one in each time step. Results from two scenarios are presented. In the first one, a team of robots explores its environment while permanently maintaining an ad hoc network structure with each other as well as a base station at a fixed location. In the second one, the robots move freely as a pack while maintaining communication with each other.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A multi-robot system can be highly beneficial for exploration, which is a core robotics task. Application domains include, for example, surveillance, reconnaissance, planetary exploration or rescue missions. When using a team of robots, the overall performance can be much faster and more robust. In this article, an approach to multi-robot exploration is presented that takes the constraints of wireless networking into account. An algorithm is introduced based on a population that samples the possible moves of all robots and a utility to select the best one in each time step. Results from two scenarios are presented. In the first one, a team of robots explores its environment while permanently maintaining an ad hoc network structure with each other as well as a base station at a fixed location. In the second one, the robots move freely as a pack while maintaining communication with each other.
Sakenas, Vytenis; Kosuchinas, Olegas; Pfingsthorn, Max; Birk, Andreas
Extraction of Semantic Floor Plans from 3D Point Cloud Maps Proceedings Article
In: IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2007.
@inproceedings{SSRR07-3D-SemanticFloorPlans,
title = {Extraction of Semantic Floor Plans from 3D Point Cloud Maps},
author = {Vytenis Sakenas and Olegas Kosuchinas and Max Pfingsthorn and Andreas Birk},
url = {https://www.researchgate.net/publication/4290695_Extraction_of_Semantic_Floor_Plans_from_3D_Point_Cloud_Maps},
doi = {10.1109/SSRR.2007.4381276},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {3D mapping is increasingly important for mobile robotics in general and for safety, security and rescue robotics (SSRR) in particular as complex environments must but captured in this domain. But it is hard to visualize 3D data in a simple way, e.g. to print maps for first responders, or to use it in standard robotics algorithms, e.g., for path planning. This paper describes a new approach to extract standard planar maps from large scale 3D maps in a very fast manner. In doing so, the approach can detect multiple floors, e.g., in a multi-story building or in a pancake collapse, and segment the 3D map accordingly. To each floor or level, a planar map is assigned, which is augmented by semantic information, especially with respect to traversability. Experiments are presented that are based on 3D maps generated in the large scale environments of USARsim, a high fidelity robot simulator. It is shown that the approach is very fast. The total processing of a complete 3D map takes just a few hundred milliseconds, leading to a proper extraction of floor plans to each of which semantic maps are assigned.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
3D mapping is increasingly important for mobile robotics in general and for safety, security and rescue robotics (SSRR) in particular as complex environments must but captured in this domain. But it is hard to visualize 3D data in a simple way, e.g. to print maps for first responders, or to use it in standard robotics algorithms, e.g., for path planning. This paper describes a new approach to extract standard planar maps from large scale 3D maps in a very fast manner. In doing so, the approach can detect multiple floors, e.g., in a multi-story building or in a pancake collapse, and segment the 3D map accordingly. To each floor or level, a planar map is assigned, which is augmented by semantic information, especially with respect to traversability. Experiments are presented that are based on 3D maps generated in the large scale environments of USARsim, a high fidelity robot simulator. It is shown that the approach is very fast. The total processing of a complete 3D map takes just a few hundred milliseconds, leading to a proper extraction of floor plans to each of which semantic maps are assigned.
Vaskevicius, Narunas; Birk, Andreas; Pathak, Kaustubh; Poppinga, Jann
Fast Detection of Polygons in 3D Point Clouds from Noise-Prone Range Sensors Proceedings Article
In: International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2007.
@inproceedings{SSRR07-3Dplanes,
title = {Fast Detection of Polygons in 3D Point Clouds from Noise-Prone Range Sensors},
author = {Narunas Vaskevicius and Andreas Birk and Kaustubh Pathak and Jann Poppinga},
url = {https://www.researchgate.net/publication/4290712_Fast_Detection_of_Polygons_in_3D_Point_Clouds_from_Noise-Prone_Range_Sensors},
doi = {10.1109/SSRR.2007.4381261},
year = {2007},
date = {2007-01-01},
urldate = {2007-01-01},
booktitle = {International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {3D sensing and modeling is increasingly important for mobile robotics in general and safety, security and rescue robotics (SSRR) in particular. To reduce the data and to allow for efficient processing, e.g., with computational geometry algorithms, it is necessary to extract surface data from 3D point clouds delivered by range sensors. A significant amount of work on this topic exists from the computer graphics community. But the existing work relies on relatively exact point cloud data. As also shown by others, sensors suited for mobile robots are very noise-prone and standard approaches that use local processing on surface normals are doomed to fail. Hence plane fitting has been suggested as solution by the robotics community. Here, a novel approach for this problem is presented. Its main feature is that it is based on region growing and that the underlying mathematics has been re-formulated such that an incremental fit can be done, i.e., the best fit surface does not have to be completely re-computed the moment a new point is investigated in the region growing process. The worst case complexity is O(n log(n)), but as shown in experiments it tends to scale linearly with typical data. Results with real world data from a Swissranger time-of-flight camera are presented where surface polygons are always successfully extracted within about 0.3 sec.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
3D sensing and modeling is increasingly important for mobile robotics in general and safety, security and rescue robotics (SSRR) in particular. To reduce the data and to allow for efficient processing, e.g., with computational geometry algorithms, it is necessary to extract surface data from 3D point clouds delivered by range sensors. A significant amount of work on this topic exists from the computer graphics community. But the existing work relies on relatively exact point cloud data. As also shown by others, sensors suited for mobile robots are very noise-prone and standard approaches that use local processing on surface normals are doomed to fail. Hence plane fitting has been suggested as solution by the robotics community. Here, a novel approach for this problem is presented. Its main feature is that it is based on region growing and that the underlying mathematics has been re-formulated such that an incremental fit can be done, i.e., the best fit surface does not have to be completely re-computed the moment a new point is investigated in the region growing process. The worst case complexity is O(n log(n)), but as shown in experiments it tends to scale linearly with typical data. Results with real world data from a Swissranger time-of-flight camera are presented where surface polygons are always successfully extracted within about 0.3 sec.
2006
Birk, Andreas; Carpin, Stefano
Merging occupancy grid maps from multiple robots Journal Article
In: IEEE Proceedings, special issue on Multi-Robot Systems, vol. 94, no. 7, pp. 1384-1397, 2006.
@article{MultiMap_IEEEproc,
title = {Merging occupancy grid maps from multiple robots},
author = {Andreas Birk and Stefano Carpin},
url = {https://www.researchgate.net/publication/2998055_Merging_Occupancy_Grid_Maps_From_Multiple_Robots},
doi = {10.1109/JPROC.2006.876965},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
journal = {IEEE Proceedings, special issue on Multi-Robot Systems},
volume = {94},
number = {7},
pages = {1384-1397},
abstract = {Mapping can potentially be speeded up in a significant way by using multiple robots exploring different parts of the environment. But the core question of multi-robot mapping is how to integrate the data of the different robots into a single global map. A significant amount of research exists in the area of multi-robot-mapping that deals with techniques to estimate the relative robots poses at the start or during the mapping process. With map merging the robots in contrast individually build local maps without any knowledge about their relative positions. The goal is then to identify regions of overlap at which the local maps can be joined together. A concrete approach to this idea is presented in form of a special similarity metric and a stochastic search algorithm. Given two maps m and m′, the search algorithm transforms m′ by rotations and translations to find a maximum overlap between m and m′. In doing so, the heuristic similarity metric guides the search algorithm toward optimal solutions. Results from experiments with up to six robots are presented based on simulated as well as real world map data. },
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Mapping can potentially be speeded up in a significant way by using multiple robots exploring different parts of the environment. But the core question of multi-robot mapping is how to integrate the data of the different robots into a single global map. A significant amount of research exists in the area of multi-robot-mapping that deals with techniques to estimate the relative robots poses at the start or during the mapping process. With map merging the robots in contrast individually build local maps without any knowledge about their relative positions. The goal is then to identify regions of overlap at which the local maps can be joined together. A concrete approach to this idea is presented in form of a special similarity metric and a stochastic search algorithm. Given two maps m and m′, the search algorithm transforms m′ by rotations and translations to find a maximum overlap between m and m′. In doing so, the heuristic similarity metric guides the search algorithm toward optimal solutions. Results from experiments with up to six robots are presented based on simulated as well as real world map data.
Birk, Andreas; Carpin, Stefano
Rescue Robotics - a crucial milestone on the road to autonomous systems Journal Article
In: Advanced Robotics Journal, vol. 20, no. 5, pp. 595-695, 2006.
@article{RescRobARJ06,
title = {Rescue Robotics - a crucial milestone on the road to autonomous systems},
author = {Andreas Birk and Stefano Carpin},
url = {https://www.researchgate.net/publication/220670453_Rescue_robotics_-_A_crucial_milestone_on_the_road_to_autonomous_systems},
doi = {10.1163/156855306776985577},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
journal = {Advanced Robotics Journal},
volume = {20},
number = {5},
pages = {595-695},
abstract = {Rescue robotics is an important steppingstone in the scientific challenge to create autonomous systems. There is a significant market for res- cue robots, which has unique features that allow a fruitful combination of application oriented developments and basic research. Unlike other markets for advanced robotics systems like service robots, the rescue robotics domain benefits from the fact that there is a human in the loop, which allows a stepwise transition from dumb teleoperated devices to truly autonomous systems. Current teleoperated devices are already very useful in this domain and they benefit from any bit of autonomy added. Human rescue workers are a scarce resource at disaster scenarios. A single operator should hence ideally supervise a multitude of robots. We present results from the rescue robots at the International University Bremen (IUB) in a core area supporting autonomy, namely mapping.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Rescue robotics is an important steppingstone in the scientific challenge to create autonomous systems. There is a significant market for res- cue robots, which has unique features that allow a fruitful combination of application oriented developments and basic research. Unlike other markets for advanced robotics systems like service robots, the rescue robotics domain benefits from the fact that there is a human in the loop, which allows a stepwise transition from dumb teleoperated devices to truly autonomous systems. Current teleoperated devices are already very useful in this domain and they benefit from any bit of autonomy added. Human rescue workers are a scarce resource at disaster scenarios. A single operator should hence ideally supervise a multitude of robots. We present results from the rescue robots at the International University Bremen (IUB) in a core area supporting autonomy, namely mapping.
Birk, Andreas; Carpin, Stefano; Chonnaparamutt, Winai; Jucikas, Viktoras; Bastani, Hamed; Delchev, Ivan; Krivulev, Ivan; Lee, Seongchu; Markov, Stefan; Pfeil, Andreas
The IUB 2005 Rescue Robot Team Book Chapter
In: Noda, Itsuki; Jacoff, Adam; Bredenfeld, Ansgar; Takahashi, Yasutake (Ed.): RoboCup 2005: Robot Soccer World Cup IX, Springer, 2006.
@inbook{rescueteam_rcup05,
title = {The IUB 2005 Rescue Robot Team},
author = {Andreas Birk and Stefano Carpin and Winai Chonnaparamutt and Viktoras Jucikas and Hamed Bastani and Ivan Delchev and Ivan Krivulev and Seongchu Lee and Stefan Markov and Andreas Pfeil},
editor = {Itsuki Noda and Adam Jacoff and Ansgar Bredenfeld and Yasutake Takahashi},
year = {2006},
date = {2006-01-01},
booktitle = {RoboCup 2005: Robot Soccer World Cup IX},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas; Condea, Cosmin
Mobile Robot Communication without the Drawbacks of Wireless Networking Book Chapter
In: Noda, Itsuki; Jacoff, Adam; Bredenfeld, Ansgar; Takahashi, Yasutake (Ed.): RoboCup 2005: Robot Soccer World Cup IX, vol. 4020, pp. 585 - 592, Springer, 2006.
@inbook{rescuecabledrum_rcup05,
title = {Mobile Robot Communication without the Drawbacks of Wireless Networking},
author = {Andreas Birk and Cosmin Condea},
editor = {Itsuki Noda and Adam Jacoff and Ansgar Bredenfeld and Yasutake Takahashi},
url = {https://www.researchgate.net/publication/220797616_Mobile_Robot_Communication_Without_the_Drawbacks_of_Wireless_Networking},
doi = {10.1007/11780519_57},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {RoboCup 2005: Robot Soccer World Cup IX},
volume = {4020},
pages = {585 - 592},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {The default solution for mobile robot communication is RF- networking, typically based on one of the IEEE 802.11 standards also known as WLAN technology. Radio communication frees the robots from umbilical cords. But it suffers from several significant drawbacks, especially limited bandwidth and range. The limitations of both aspects are in addition hard to predict as they are strongly dependent on environment conditions. An outdoor RF-link may easily cover 100m over a line-of-sight with full bandwidth. In an indoor environment, the range often drops to a few rooms. Walls made of hardened concrete even completely block the communication. Driven by a concrete application scenario where communication is vital, namely robot rescue, we developed a communication system based on glassfibre links. The system provides 100MBit ethernet connections over up to 100m in its default configuration. The glassfibres provide high bandwidth, they are very lightweight and thin, and they can take a lot of stress, much more than normal copper cable. The glassfiber links are deployed from the mobile robot via a cable drum. The system is based on media converters at both ends. One of them is integrated on the drum, thus allowing the usage of inexpensive wired sliprings. The glassfibre system turned out to be very performant and reliable, both in operation in the challenging environment of rescue robotics as well as in concrete experiments.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The default solution for mobile robot communication is RF- networking, typically based on one of the IEEE 802.11 standards also known as WLAN technology. Radio communication frees the robots from umbilical cords. But it suffers from several significant drawbacks, especially limited bandwidth and range. The limitations of both aspects are in addition hard to predict as they are strongly dependent on environment conditions. An outdoor RF-link may easily cover 100m over a line-of-sight with full bandwidth. In an indoor environment, the range often drops to a few rooms. Walls made of hardened concrete even completely block the communication. Driven by a concrete application scenario where communication is vital, namely robot rescue, we developed a communication system based on glassfibre links. The system provides 100MBit ethernet connections over up to 100m in its default configuration. The glassfibres provide high bandwidth, they are very lightweight and thin, and they can take a lot of stress, much more than normal copper cable. The glassfiber links are deployed from the mobile robot via a cable drum. The system is based on media converters at both ends. One of them is integrated on the drum, thus allowing the usage of inexpensive wired sliprings. The glassfibre system turned out to be very performant and reliable, both in operation in the challenging environment of rescue robotics as well as in concrete experiments.
Birk, Andreas; Markov, Stefan; Delchev, Ivan; Pathak, Kaustubh
Autonomous Rescue Operations on the IUB Rugbot Book Chapter
In: IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2006.
@inbook{autonomy_ssrr06,
title = {Autonomous Rescue Operations on the IUB Rugbot},
author = {Andreas Birk and Stefan Markov and Ivan Delchev and Kaustubh Pathak},
url = {https://www.researchgate.net/publication/250234379_Autonomous_Rescue_Operations_on_the_IUB_Rugbot},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {Full autonomy is an interesting feature for rescue robots not only from a basic research perspective but also in respect to an application oriented view. Autonomy can help a single operator to handle multiple robots or to manage areas that suffer from disturbances in the RF-communication. Furthermore, the different modules used for autonomy like mapping and victim detection are also tremendously helpful for tele-operation. In this paper, an autonomy system is described that was successfully tested on the IUB rescue robots in the lab as well as on the RoboCup Dutch Open 2006 competition. The basic system features all core modules for autonomy. Furthermore, we describe a novel way of victim detection. },
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Full autonomy is an interesting feature for rescue robots not only from a basic research perspective but also in respect to an application oriented view. Autonomy can help a single operator to handle multiple robots or to manage areas that suffer from disturbances in the RF-communication. Furthermore, the different modules used for autonomy like mapping and victim detection are also tremendously helpful for tele-operation. In this paper, an autonomy system is described that was successfully tested on the IUB rescue robots in the lab as well as on the RoboCup Dutch Open 2006 competition. The basic system features all core modules for autonomy. Furthermore, we describe a novel way of victim detection.
Birk, Andreas; Pathak, Kausthub; Schwertfeger, Soeren; Chonnaparamutt, Winai
The IUB Rugbot: an intelligent, rugged mobile robot for search and rescue operations Book Chapter
In: IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR), IEEE Press, 2006.
@inbook{rugbot_ssrr06,
title = {The IUB Rugbot: an intelligent, rugged mobile robot for search and rescue operations},
author = {Andreas Birk and Kausthub Pathak and Soeren Schwertfeger and Winai Chonnaparamutt},
url = {https://www.researchgate.net/publication/250206089_The_IUB_Rugbot_an_intelligent_rugged_mobile_robot_for_search_and_rescue_operations},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {IEEE International Workshop on Safety, Security, and Rescue Robotics (SSRR)},
publisher = {IEEE Press},
abstract = {The paper describes the IUB Rugbot, a rugged mobile robot that features quite some on-board intelligence. The robot and its software is the latest development of the IUB rescue robot team, which is active since 2001 in this research area. IUB robotics takes an integrated approach to rescue robots. This means that the development of the according systems ranges from the basic mechatronics to the high-level functionalities for intelligent behavior.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The paper describes the IUB Rugbot, a rugged mobile robot that features quite some on-board intelligence. The robot and its software is the latest development of the IUB rescue robot team, which is active since 2001 in this research area. IUB robotics takes an integrated approach to rescue robots. This means that the development of the according systems ranges from the basic mechatronics to the high-level functionalities for intelligent behavior.
Birk, Andreas; Pfingsthorn, Max
A HMI supporting Adjustable Autonomy of Rescue Robots Book Chapter
In: Noda, Itsuki; Jacoff, Adam; Bredenfeld, Ansgar; Takahashi, Yasutake (Ed.): RoboCup 2005: Robot Soccer World Cup IX, vol. 4020, pp. 255-266, Springer, 2006.
@inbook{rescueGUI_rcup05,
title = {A HMI supporting Adjustable Autonomy of Rescue Robots},
author = {Andreas Birk and Max Pfingsthorn},
editor = {Itsuki Noda and Adam Jacoff and Ansgar Bredenfeld and Yasutake Takahashi},
url = {https://www.researchgate.net/publication/220797332_A_HMI_Supporting_Adjustable_Autonomy_of_Rescue_Robots},
doi = {10.1007/11780519_23},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {RoboCup 2005: Robot Soccer World Cup IX},
volume = {4020},
pages = {255-266},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Human rescue workers are a scarce resource at disaster sites. But it is still a long way to go before fully autonomous rescue robots will be fieldable. The usefulness of rescue robots will hence strongly depend on the availability of user interfaces that enable a single first responder to operate a whole set of robots. For this challenge, it is important to preprocess and streamline the immense data flow from the robots and to assist the operator as much as possible in the processes of controlling the robots. This paper introduces an adaptive graphical interface supporting adjustable autonomy of rescue robots. The design is based on insights from the literature in this field where intensive surveys of the actual needs in this domain were compiled.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Human rescue workers are a scarce resource at disaster sites. But it is still a long way to go before fully autonomous rescue robots will be fieldable. The usefulness of rescue robots will hence strongly depend on the availability of user interfaces that enable a single first responder to operate a whole set of robots. For this challenge, it is important to preprocess and streamline the immense data flow from the robots and to assist the operator as much as possible in the processes of controlling the robots. This paper introduces an adaptive graphical interface supporting adjustable autonomy of rescue robots. The design is based on insights from the literature in this field where intensive surveys of the actual needs in this domain were compiled.
Carpin, Stefano; Birk, Andreas; Lewis, Michael; Jacoff, Adam
High fidelity tools for rescue robotics: results and perspectives Book Chapter
In: Noda, Itsuki; Jacoff, Adam; Bredenfeld, Ansgar; Takahashi, Yasutake (Ed.): RoboCup 2005: Robot Soccer World Cup IX, Springer, 2006.
@inbook{virtualrobot_rcup05,
title = {High fidelity tools for rescue robotics: results and perspectives},
author = {Stefano Carpin and Andreas Birk and Michael Lewis and Adam Jacoff},
editor = {Itsuki Noda and Adam Jacoff and Ansgar Bredenfeld and Yasutake Takahashi},
url = {https://www.researchgate.net/publication/220797334_High_Fidelity_Tools_for_Rescue_Robotics_Results_and_Perspectives},
doi = {10.1007/11780519_27},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {RoboCup 2005: Robot Soccer World Cup IX},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {USARSim is a high fidelity robot simulation tool based on a commercial game engine. We illustrate the overall structure of the simulator and we argue about its use as a bridging tool between the RoboCupRescue Real Robot League and the RoboCupRescue Simulation League. In particular we show some results concerning the validation of the system. Algorithms useful for the search and rescue task have been developed in the simulator and then executed on real robots providing encouraging results.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
USARSim is a high fidelity robot simulation tool based on a commercial game engine. We illustrate the overall structure of the simulator and we argue about its use as a bridging tool between the RoboCupRescue Real Robot League and the RoboCupRescue Simulation League. In particular we show some results concerning the validation of the system. Algorithms useful for the search and rescue task have been developed in the simulator and then executed on real robots providing encouraging results.
Chonnaparamutt, Winai; Birk, Andreas
Using Rescue Robots to Increase Construction Site Safety Proceedings Article
In: 23rd International Symposium on Automation and Robotics in Construction (ISARC), 2006.
@inproceedings{ConstructionSiteSafety,
title = {Using Rescue Robots to Increase Construction Site Safety},
author = {Winai Chonnaparamutt and Andreas Birk},
url = {https://www.researchgate.net/publication/320562548_Using_Rescue_Robots_to_Increase_Construction_Site_Safety},
doi = {10.22260/ISARC2006/0047},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {23rd International Symposium on Automation and Robotics in Construction (ISARC)},
abstract = {Robotics technology is continuously gaining in importance in many applications. Especially mobile robots are already widely used in many fields including construction sites. Accidents in the construction industry have been identified as one of the most hazardous incidents compared to other work environments. We therefore propose to consider robotics technology to improve this situation. This paper introduces the idea of applying rescue robots as an accident prevention and first aid provision system at construction sites. In doing so, the mobility of the robot is a key issue. The paper presents a special robot, the rugbot, and a novel mechatronic component, a so-called flipper, which can be used to adapt the footprint of the robot. This allows the robot to negotiate a difficult environment as it can be found at construction sites.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Robotics technology is continuously gaining in importance in many applications. Especially mobile robots are already widely used in many fields including construction sites. Accidents in the construction industry have been identified as one of the most hazardous incidents compared to other work environments. We therefore propose to consider robotics technology to improve this situation. This paper introduces the idea of applying rescue robots as an accident prevention and first aid provision system at construction sites. In doing so, the mobility of the robot is a key issue. The paper presents a special robot, the rugbot, and a novel mechatronic component, a so-called flipper, which can be used to adapt the footprint of the robot. This allows the robot to negotiate a difficult environment as it can be found at construction sites.
Rooker, Martijn; Birk, Andreas
Communicative Exploration with Robot Packs Book Chapter
In: Noda, Itsuki; Jacoff, Adam; Bredenfeld, Ansgar; Takahashi, Yasutake (Ed.): RoboCup 2005: Robot Soccer World Cup IX, vol. 4020, pp. 267 - 278, Springer, 2006.
@inbook{commexplore_rcup05,
title = {Communicative Exploration with Robot Packs},
author = {Martijn Rooker and Andreas Birk},
editor = {Itsuki Noda and Adam Jacoff and Ansgar Bredenfeld and Yasutake Takahashi},
url = {https://www.researchgate.net/publication/220797498_Communicative_Exploration_with_Robot_Packs},
doi = {10.1007/11780519_24},
year = {2006},
date = {2006-01-01},
urldate = {2006-01-01},
booktitle = {RoboCup 2005: Robot Soccer World Cup IX},
volume = {4020},
pages = {267 - 278},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Exploration is a core challenge for RoboCup Rescue. So-called communicative exploration is a novel strategy for multi-robot exploration that unlike other approaches takes the limits of wireless communication systems into account. Here, previous results that where achieved for a team of robots linked to a basestation are significantly extended to also cover robot packs, i.e., multi-robot teams that are not permanently tied to an operator's station. Unlike teams that are con- strained by the immobility of a basestation, packs can explore arbitrarily large regions. Results from experiments with packs of 4, 5 and 6 robots are presented. The new strategy constantly maintains the communication between the robots while exploring, whereas the commonly used frontier-based exploration strategy, which is used in the experiments as comparison to our approach, leads to a rapid loss of communication.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Exploration is a core challenge for RoboCup Rescue. So-called communicative exploration is a novel strategy for multi-robot exploration that unlike other approaches takes the limits of wireless communication systems into account. Here, previous results that where achieved for a team of robots linked to a basestation are significantly extended to also cover robot packs, i.e., multi-robot teams that are not permanently tied to an operator's station. Unlike teams that are con- strained by the immobility of a basestation, packs can explore arbitrarily large regions. Results from experiments with packs of 4, 5 and 6 robots are presented. The new strategy constantly maintains the communication between the robots while exploring, whereas the commonly used frontier-based exploration strategy, which is used in the experiments as comparison to our approach, leads to a rapid loss of communication.
Steels, Luc; Birk, Andreas
Robotics Book Chapter
In: Encyclopedia of Cognitive Science, Wiley, 2006.
@inbook{steels_birk_robotics06,
title = {Robotics},
author = {Luc Steels and Andreas Birk},
doi = {https://doi.org/10.1002/0470018860.s00073},
year = {2006},
date = {2006-01-01},
booktitle = {Encyclopedia of Cognitive Science},
publisher = {Wiley},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
2005
Birk, Andreas
The IUB 2004 Rescue Robot Team Book Chapter
In: Nardi, Daniele; Riedmiller, Martin; Sammut, Claude (Ed.): RoboCup 2004: Robot Soccer World Cup VIII, vol. 3276, Springer, 2005.
@inbook{rescueteam_rcup04,
title = {The IUB 2004 Rescue Robot Team},
author = {Andreas Birk},
editor = {Daniele Nardi and Martin Riedmiller and Claude Sammut},
year = {2005},
date = {2005-01-01},
booktitle = {RoboCup 2004: Robot Soccer World Cup VIII},
volume = {3276},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Carpin, Stefano; Birk, Andreas
Stochastic map merging in rescue environments Book Chapter
In: Nardi, Daniele; Riedmiller, Martin; Sammut, Claude (Ed.): RoboCup 2004: Robot Soccer World Cup VIII, vol. 3276, pp. p.483ff, Springer, 2005.
@inbook{multimap_rcup04,
title = {Stochastic map merging in rescue environments},
author = {Stefano Carpin and Andreas Birk},
editor = {Daniele Nardi and Martin Riedmiller and Claude Sammut},
year = {2005},
date = {2005-01-01},
booktitle = {RoboCup 2004: Robot Soccer World Cup VIII},
volume = {3276},
pages = {p.483ff},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Carpin, Stefano; Birk, Andreas; Jucikas, Victoras
On Map Merging Journal Article
In: International Journal of Robotics and Autonomous Systems, vol. 53, pp. 1–14, 2005.
@article{multimap_ras05,
title = {On Map Merging},
author = {Stefano Carpin and Andreas Birk and Victoras Jucikas},
url = {https://www.researchgate.net/publication/222668559_On_map_merging},
doi = {10.1016/j.robot.2005.07.001},
year = {2005},
date = {2005-01-01},
urldate = {2005-01-01},
journal = {International Journal of Robotics and Autonomous Systems},
volume = {53},
pages = {1--14},
abstract = {We illustrate our experience in developing and implementing algorithms for map merging, i.e., the problem of fusing two or more partial maps without common reference frames into one large global map. The partial maps may for example be acquired by multiple robots, or during several runs of a single robot from varying starting positions. Our work deals with low quality maps based on probabilistic grids, motivated by the goal to develop multiple mobile platforms to be used in rescue environments. Several contributions to map merging are presented. First of all, we address map merging using a motion planning algorithm. The merging process can be done by rotating and translating the partial maps until similar regions overlap. Second, a motion planning algorithm is presented which is particular suited for this task. Third, a special metric is presented which guides the motion planning algorithm towards the goal of optimally overlapping partial maps. Results with our approach are presented based on data gathered from real robots developed for the RoboCup Rescue real robot league.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
We illustrate our experience in developing and implementing algorithms for map merging, i.e., the problem of fusing two or more partial maps without common reference frames into one large global map. The partial maps may for example be acquired by multiple robots, or during several runs of a single robot from varying starting positions. Our work deals with low quality maps based on probabilistic grids, motivated by the goal to develop multiple mobile platforms to be used in rescue environments. Several contributions to map merging are presented. First of all, we address map merging using a motion planning algorithm. The merging process can be done by rotating and translating the partial maps until similar regions overlap. Second, a motion planning algorithm is presented which is particular suited for this task. Third, a special metric is presented which guides the motion planning algorithm towards the goal of optimally overlapping partial maps. Results with our approach are presented based on data gathered from real robots developed for the RoboCup Rescue real robot league.
Kenn, Holger; Birk, Andreas
From Games to Applications: Component reuse in Rescue Robots Book Chapter
In: Nardi, Daniele; Riedmiller, Martin; Sammut, Claude (Ed.): RoboCup 2004: Robot Soccer World Cup VIII, vol. 3276, pp. p.669ff, Springer, 2005.
@inbook{cuberescue_rcup04,
title = {From Games to Applications: Component reuse in Rescue Robots},
author = {Holger Kenn and Andreas Birk},
editor = {Daniele Nardi and Martin Riedmiller and Claude Sammut},
url = {https://www.researchgate.net/publication/220797794_From_Games_to_Applications_Component_Reuse_in_Rescue_Robots},
doi = {10.1007/978-3-540-32256-6_65},
year = {2005},
date = {2005-01-01},
urldate = {2005-01-01},
booktitle = {RoboCup 2004: Robot Soccer World Cup VIII},
volume = {3276},
pages = {p.669ff},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {Component-based software engineering is useful for embedded applications such as robotics. However, heavyweight component systems such as CORBA overstrain the resources available in many embedded systems. Here, a lightweight component-based approach is used to implement the system software of the so-called CubeSystem, CubeOS. Since 1998, CubeOS and its component system have been successfully used in various areas from industry projects over RoboCup-related research to edutainment applications. Many of the components used in RoboCup soccer have been carried over in the implementation of the IUB Rescue robots, demonstrating the potential for software reuse.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Component-based software engineering is useful for embedded applications such as robotics. However, heavyweight component systems such as CORBA overstrain the resources available in many embedded systems. Here, a lightweight component-based approach is used to implement the system software of the so-called CubeSystem, CubeOS. Since 1998, CubeOS and its component system have been successfully used in various areas from industry projects over RoboCup-related research to edutainment applications. Many of the components used in RoboCup soccer have been carried over in the implementation of the IUB Rescue robots, demonstrating the potential for software reuse.
Rooker, Martijn; Birk, Andreas
Combining Exploration and Ad-Hoc Networking in RoboCup Rescue Book Chapter
In: Nardi, Daniele; Riedmiller, Martin; Sammut, Claude (Ed.): RoboCup 2004: Robot Soccer World Cup VIII, vol. 3276, pp. pp.236-246, Springer, 2005.
@inbook{commexlpore_rcup04,
title = {Combining Exploration and Ad-Hoc Networking in RoboCup Rescue},
author = {Martijn Rooker and Andreas Birk},
editor = {Daniele Nardi and Martin Riedmiller and Claude Sammut},
url = {https://www.researchgate.net/publication/220797746_Combining_Exploration_and_Ad-Hoc_Networking_in_RoboCup_Rescue},
doi = {10.1007/978-3-540-32256-6_19},
year = {2005},
date = {2005-01-01},
urldate = {2005-01-01},
booktitle = {RoboCup 2004: Robot Soccer World Cup VIII},
volume = {3276},
pages = {pp.236-246},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
abstract = {In challenging environments where the risk of loss of a robot is high, robot teams are a natural choice. In many applications like for example rescue missions there are two crucial tasks for the robots. First, they have to efficiently and exhaustively explore the environment. Second, they must keep up a network connection to the base-station to transmit data to ensure timely arrival and secure storage of vital information. When using wireless media, it is necessary to use robots from the team as relay stations for this purpose. This paper deals with the problem to combine an efficient exploration of the environment with suited motions of the robots to keep data transmissions stable.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
In challenging environments where the risk of loss of a robot is high, robot teams are a natural choice. In many applications like for example rescue missions there are two crucial tasks for the robots. First, they have to efficiently and exhaustively explore the environment. Second, they must keep up a network connection to the base-station to transmit data to ensure timely arrival and secure storage of vital information. When using wireless media, it is necessary to use robots from the team as relay stations for this purpose. This paper deals with the problem to combine an efficient exploration of the environment with suited motions of the robots to keep data transmissions stable.
2004
Birk, Andreas
The Evolution of Cooperation in Groups with up to 25 Agents Proceedings Article
In: Artificial Intelligence and Applications, AIA'04, ACTA Press, 2004.
@inproceedings{birk_AIA04_coop_evol,
title = {The Evolution of Cooperation in Groups with up to 25 Agents},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/374826262_The_Evolution_of_Cooperation_in_Groups_with_up_to_25_Agents},
year = {2004},
date = {2004-01-01},
urldate = {2004-01-01},
booktitle = {Artificial Intelligence and Applications, AIA'04},
publisher = {ACTA Press},
abstract = {The iterated prisoner's dilemma (iPD) in its standard form is a very well known, popular basis for research on cooperation through evolutionary game theory. Based on a generalization of the iPD in form of a N-players game with continuous degrees of cooperation, the novel strategy of so-called justified-snobism (JS) is presented. This strategy tries to cooperate slightly more than the average of the group of players. The results presented in this paper extends previous work as here cooperation is achieved in relatively large groups with N = 25 players.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The iterated prisoner's dilemma (iPD) in its standard form is a very well known, popular basis for research on cooperation through evolutionary game theory. Based on a generalization of the iPD in form of a N-players game with continuous degrees of cooperation, the novel strategy of so-called justified-snobism (JS) is presented. This strategy tries to cooperate slightly more than the average of the group of players. The results presented in this paper extends previous work as here cooperation is achieved in relatively large groups with N = 25 players.
Birk, Andreas
Fast Robot Prototyping with the CubeSystem Proceedings Article
In: Proceedings of the International Conference on Robotics and Automation, IEEE Press, 2004.
@inproceedings{birk_cube_icra04,
title = {Fast Robot Prototyping with the CubeSystem},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/224748805_Fast_robot_prototyping_with_the_CubeSystem},
doi = {10.1109/ROBOT.2004.1302539},
year = {2004},
date = {2004-01-01},
urldate = {2004-01-01},
booktitle = {Proceedings of the International Conference on Robotics and Automation},
publisher = {IEEE Press},
abstract = {The CubeSystem is a collection of hardware- and software-components for fast robot prototyping. The main goal of the CubeSystem project is to provide an open source collection of generic building blocks that can be freely combined into an application. This paper describes the first release of the CubeSystem, that evolved in more than five years of research and development. The benefits of the CubeSystem are illustrated by several applications, ranging from educational activities to industrial projects.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
The CubeSystem is a collection of hardware- and software-components for fast robot prototyping. The main goal of the CubeSystem project is to provide an open source collection of generic building blocks that can be freely combined into an application. This paper describes the first release of the CubeSystem, that evolved in more than five years of research and development. The benefits of the CubeSystem are illustrated by several applications, ranging from educational activities to industrial projects.
Birk, Andreas
The IUB Rescue Arena, a testbed for rescue robots research Book Chapter
In: Second IEEE International Workshop on Safety, Security, and Rescue Robotics, SSRR'04, 2004.
@inbook{birk_arena04,
title = {The IUB Rescue Arena, a testbed for rescue robots research},
author = {Andreas Birk},
year = {2004},
date = {2004-01-01},
booktitle = {Second IEEE International Workshop on Safety, Security, and Rescue Robotics, SSRR'04},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas
Learning to Cooperate via a Selectionist Algorithm Proceedings Article
In: Artificial Intelligence and Applications, AIA'04, ACTA Press, 2004.
@inproceedings{birk_AIA04_coop_learning,
title = {Learning to Cooperate via a Selectionist Algorithm},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/374834827_Learning_to_Cooperate_via_a_Selectionist_Algorithm},
year = {2004},
date = {2004-01-01},
urldate = {2004-01-01},
booktitle = {Artificial Intelligence and Applications, AIA'04},
publisher = {ACTA Press},
abstract = {Evolutionary game-theory is a popular way to investigate models of cooperation. But this has the obvious disadvantage that evolution with the propagation of genes over generations is an unrealistic assumption to investigate fast-changing social interactions. In this paper, it is shown how a transition from evolutionary game-theory to a learning model can be made. Especially, results from experiments in an N-Players Prisoners' Dilemma are presented where the agents learn to cooperate despite significant temptations to cheat.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Evolutionary game-theory is a popular way to investigate models of cooperation. But this has the obvious disadvantage that evolution with the propagation of genes over generations is an unrealistic assumption to investigate fast-changing social interactions. In this paper, it is shown how a transition from evolutionary game-theory to a learning model can be made. Especially, results from experiments in an N-Players Prisoners' Dilemma are presented where the agents learn to cooperate despite significant temptations to cheat.
Birk, Andreas; Carpin, Stefano; Kenn, Holger
The IUB 2003 Rescue Robot Team Book Chapter
In: Polani, D.; Browning, B.; Bonarini, A.; Yoshida, K. (Ed.): RoboCup 2003: Robot Soccer World Cup VII, vol. 3020, Springer, 2004.
@inbook{rescueteam_rcup03,
title = {The IUB 2003 Rescue Robot Team},
author = {Andreas Birk and Stefano Carpin and Holger Kenn},
editor = {D. Polani and B. Browning and A. Bonarini and K. Yoshida},
year = {2004},
date = {2004-01-01},
booktitle = {RoboCup 2003: Robot Soccer World Cup VII},
volume = {3020},
publisher = {Springer},
series = {Lecture Notes in Artificial Intelligence (LNAI)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Carpin, Stefano; Jucikas, Victoras; Birk, Andreas
Multi-robot mapping for rescue robotics Proceedings Article
In: Proceedings of the 2004 International Workshop on Safety, Security and Rescue Robotics (SSRR), 2004.
@inproceedings{CarpinSSRR,
title = {Multi-robot mapping for rescue robotics},
author = {Stefano Carpin and Victoras Jucikas and Andreas Birk},
url = {https://www.researchgate.net/publication/374819932_Multi-robot_mapping_for_rescue_robotics},
year = {2004},
date = {2004-01-01},
urldate = {2004-01-01},
booktitle = {Proceedings of the 2004 International Workshop on Safety, Security and Rescue Robotics (SSRR)},
abstract = {We illustrate our progresses in developing multi-robot systems to be used for mapping in rescue scenarios. The problem we are currently investigating is to combine poor quality multiple maps produced by different robots into a single map to be used by human operators. In particular we motivate our approach and we illustrate the different techniques we implemented and that are at the moment being compared.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
We illustrate our progresses in developing multi-robot systems to be used for mapping in rescue scenarios. The problem we are currently investigating is to combine poor quality multiple maps produced by different robots into a single map to be used by human operators. In particular we motivate our approach and we illustrate the different techniques we implemented and that are at the moment being compared.
Rooker, Martijn; Birk, Andreas
Communicative Exploration in Dangerous Environments Proceedings Article
In: Second International Workshop on Advances in Service Robotics (ASER'04), 2004.
@inproceedings{aser04,
title = {Communicative Exploration in Dangerous Environments},
author = {Martijn Rooker and Andreas Birk},
url = {https://www.researchgate.net/publication/274891451_Communicative_Exploration_in_Dangerous_Environments},
year = {2004},
date = {2004-01-01},
urldate = {2004-01-01},
booktitle = {Second International Workshop on Advances in Service Robotics (ASER'04)},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
2003
Birk, Andreas
The Evolution of Cooperation in Groups with up to 25 Agents Book Chapter
In: The Third IASTED International Conference on Artificial Intelligence and Applications, AIA'03, 2003.
@inbook{birk_aia03_evol,
title = {The Evolution of Cooperation in Groups with up to 25 Agents},
author = {Andreas Birk},
url = {https://www.researchgate.net/publication/374826262_The_Evolution_of_Cooperation_in_Groups_with_up_to_25_Agents},
year = {2003},
date = {2003-01-01},
urldate = {2003-01-01},
booktitle = {The Third IASTED International Conference on Artificial Intelligence and Applications, AIA'03},
abstract = {The iterated prisoner's dilemma (iPD) in its standard form is a very well known, popular basis for research on cooperation through evolutionary game theory. Based on a generalization of the iPD in form of a N-players game with continuous degrees of cooperation, the novel strategy of so-called justified-snobism (JS) is presented. This strategy tries to cooperate slightly more than the average of the group of players. The results presented in this paper extends previous work as here cooperation is achieved in relatively large groups with N = 25 players.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The iterated prisoner's dilemma (iPD) in its standard form is a very well known, popular basis for research on cooperation through evolutionary game theory. Based on a generalization of the iPD in form of a N-players game with continuous degrees of cooperation, the novel strategy of so-called justified-snobism (JS) is presented. This strategy tries to cooperate slightly more than the average of the group of players. The results presented in this paper extends previous work as here cooperation is achieved in relatively large groups with N = 25 players.
Birk, A.; Carpin, S.; Kenn, H.
The IUB 2003 Rescue Robot Team Book Chapter
In: Robocup 2003, Team Description Paper (TDP), 2003.
@inbook{TDP2003,
title = {The IUB 2003 Rescue Robot Team},
author = {A. Birk and S. Carpin and H. Kenn},
year = {2003},
date = {2003-01-01},
booktitle = {Robocup 2003},
publisher = {Team Description Paper (TDP)},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas; Kenn, Holger
A Rescue Robot Control Architecture for a Rescue Robot ensuring Safe Semi-Autonomous Operation Book Chapter
In: Kaminka, Gal; Lima, Pedro U.; Rojas, Raul (Ed.): RoboCup-02: Robot Soccer World Cup VI, vol. 2752, pp. 254-262, Springer, 2003.
@inbook{rescuearchitecture_rcup02,
title = {A Rescue Robot Control Architecture for a Rescue Robot ensuring Safe Semi-Autonomous Operation},
author = {Andreas Birk and Holger Kenn},
editor = {Gal Kaminka and Pedro U. Lima and Raul Rojas},
url = {https://www.researchgate.net/publication/220797287_A_Rescue_Robot_Control_Architecture_Ensuring_Safe_Semi-autonomous_Operation},
doi = {10.1007/978-3-540-45135-8_19},
year = {2003},
date = {2003-01-01},
urldate = {2003-01-01},
booktitle = {RoboCup-02: Robot Soccer World Cup VI},
volume = {2752},
pages = {254-262},
publisher = {Springer},
series = {LNAI},
abstract = {The rescue robots developed at the International University Bremen (IUB) are semi-autonomous mobile robots providing streams of video and other essential data via wireless connections to human operated basestations, supplemented by various basic and optional behaviors on board of the robots. Due to the limitations of wireless connections and the complexity of rescue operations, the full operation of a robot can not be constantly supervised by a human operator, i.e., the robots have to be semi-autonomous. This paper describes how the main challenge of safe operation under semi-autonomous control can in general be solved. The key elements are a special software architecture and a scheduling framework that ensure Quality of Service (QoS) and Fail-Safe Guarantees (FSG) despite the unpredictable performance of standard Internet/Intranet-technologies, especially when wireless components are involved.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
The rescue robots developed at the International University Bremen (IUB) are semi-autonomous mobile robots providing streams of video and other essential data via wireless connections to human operated basestations, supplemented by various basic and optional behaviors on board of the robots. Due to the limitations of wireless connections and the complexity of rescue operations, the full operation of a robot can not be constantly supervised by a human operator, i.e., the robots have to be semi-autonomous. This paper describes how the main challenge of safe operation under semi-autonomous control can in general be solved. The key elements are a special software architecture and a scheduling framework that ensure Quality of Service (QoS) and Fail-Safe Guarantees (FSG) despite the unpredictable performance of standard Internet/Intranet-technologies, especially when wireless components are involved.
Birk, Andreas; Kenn, Holger; Carpin, Stefano; Pfingsthorn, Max
Toward Autonomous Rescue Robots Proceedings Article
In: Proceedings of the First International Workshop on Synthetic Simulation and Robotics to Mitigate Earthquake Disasters, Padova, Italy, 2003.
@inproceedings{pfingsthorn03autonomous,
title = {Toward Autonomous Rescue Robots},
author = {Andreas Birk and Holger Kenn and Stefano Carpin and Max Pfingsthorn},
year = {2003},
date = {2003-01-01},
urldate = {2003-01-01},
booktitle = {Proceedings of the First International Workshop on Synthetic Simulation and Robotics to Mitigate Earthquake Disasters},
address = {Padova, Italy},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
Birk, Andreas; Kenn, Holger; Pfingsthorn, Max
The IUB Rescue Robots: From Webcams to Lifesavers Book Chapter
In: Proceedings of the ASER '03, 1st International Workshop on Advances in Service Robotics, 2003.
@inbook{pfingsthorn03lifesavers,
title = {The IUB Rescue Robots: From Webcams to Lifesavers},
author = {Andreas Birk and Holger Kenn and Max Pfingsthorn},
url = {https://www.researchgate.net/publication/374470930_The_IUB_Rescue_Robots_From_Webcams_to_Lifesavers},
year = {2003},
date = {2003-01-01},
urldate = {2003-01-01},
booktitle = {Proceedings of the ASER '03, 1st International Workshop on Advances in Service Robotics},
abstract = {Mobile robots can be highly valuable tools in Urban Rescue Missions, inspecting collapsed buildings, assessing the situation, looking for victims and establishing a first contact to them. At the International University Bremen (IUB), rescue robots for these tasks are developed. The American National Institute for Standards and Technology (NIST) has developed a special testing arena based on data from real catastrophes. This paper describes the IUB rescue robots that have been tested in the NIST arena during RoboCup 2002 in Fukuoka, Japan. The robots are semi-autonomous mobile robots providing streams of video and other essential data via wireless connections to human operated base-stations. All components of the robots, ranging from mechanics over computational hardware to all software levels have been to a large extent custom designed for the difficult task of rescue missions. This paper gives a short overview of all components, but it focuses especially on the implementation of the data streams from and to the robots.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Mobile robots can be highly valuable tools in Urban Rescue Missions, inspecting collapsed buildings, assessing the situation, looking for victims and establishing a first contact to them. At the International University Bremen (IUB), rescue robots for these tasks are developed. The American National Institute for Standards and Technology (NIST) has developed a special testing arena based on data from real catastrophes. This paper describes the IUB rescue robots that have been tested in the NIST arena during RoboCup 2002 in Fukuoka, Japan. The robots are semi-autonomous mobile robots providing streams of video and other essential data via wireless connections to human operated base-stations. All components of the robots, ranging from mechanics over computational hardware to all software levels have been to a large extent custom designed for the difficult task of rescue missions. This paper gives a short overview of all components, but it focuses especially on the implementation of the data streams from and to the robots.
Carpin, S.; Kenn, H.; Birk, A.
Autonomous Mapping in the Real Robot Rescue League Book Chapter
In: Robocup 2003, Springer, 2003.
@inbook{CARPIN03,
title = {Autonomous Mapping in the Real Robot Rescue League},
author = {S. Carpin and H. Kenn and A. Birk},
year = {2003},
date = {2003-01-01},
booktitle = {Robocup 2003},
publisher = {Springer},
series = {LNAI},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Kenn, Holger; Carpin, Stefano; Pfingsthorn, Max; Liebald, Benjamin; Hepes, Ioan; Ciocov, Catalin; Birk, Andreas
FAST-Robotics: a rapid-prototyping framework for intelligent mobile robotics Book Chapter
In: Proceedings of the 2003 IASTED International Conference on Artificial Intelligence and Applications, pp. 76-81, Benalmadena, Malaga, Spain, 2003.
@inbook{pfingsthorn03fast,
title = {FAST-Robotics: a rapid-prototyping framework for intelligent mobile robotics},
author = {Holger Kenn and Stefano Carpin and Max Pfingsthorn and Benjamin Liebald and Ioan Hepes and Catalin Ciocov and Andreas Birk},
url = {https://www.researchgate.net/publication/228605146_FAST-robots_A_rapid_prototyping_framework_for_intelligent_mobile_robotics},
year = {2003},
date = {2003-01-01},
urldate = {2003-01-01},
booktitle = {Proceedings of the 2003 IASTED International Conference on Artificial Intelligence and Applications},
pages = {76-81},
address = {Benalmadena, Malaga, Spain},
abstract = {FAST-Robots (Framework Architecture for Self-controlled and Teleoperated Robots) is an object-oriented network control architecture framework for robotics experiments that supports mixed teleoperation and autonomy. Its features are the abstraction from the underlying communication system and its simplicity and flexibility towards modification and extension. This architecture is used aboard mobile robots in the Robocup Rescue Robots League.},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
FAST-Robots (Framework Architecture for Self-controlled and Teleoperated Robots) is an object-oriented network control architecture framework for robotics experiments that supports mixed teleoperation and autonomy. Its features are the abstraction from the underlying communication system and its simplicity and flexibility towards modification and extension. This architecture is used aboard mobile robots in the Robocup Rescue Robots League.
2002
Birk, Andreas; Kenn, Holger; Rooker, Martijn; Akhil, Agrawal; Vlad, Balan Horia; Nina, Burger; Christoph, Burger-Scheidlin; Vinod, Devanathan; Dumitru, Erhan; Ioan, Hepes; Aakash, Jain; Premvir, Jain; Benjamin, Liebald; Ge, Luksys
The IUB 2002 Rescue Robot Team Book Chapter
In: Kaminka, Gal; Lima, Pedro U.; Rojas, Raul (Ed.): RoboCup-02: Robot Soccer World Cup VI, Springer, 2002.
@inbook{rescueteam_rcup02,
title = {The IUB 2002 Rescue Robot Team},
author = {Andreas Birk and Holger Kenn and Martijn Rooker and Agrawal Akhil and Balan Horia Vlad and Burger Nina and Burger-Scheidlin Christoph and Devanathan Vinod and Erhan Dumitru and Hepes Ioan and Jain Aakash and Jain Premvir and Liebald Benjamin and Luksys Ge},
editor = {Gal Kaminka and Pedro U. Lima and Raul Rojas},
year = {2002},
date = {2002-01-01},
booktitle = {RoboCup-02: Robot Soccer World Cup VI},
publisher = {Springer},
series = {LNAI},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas; Kenn, Holger; Rooker, Martijn; Akhil, Agrawal; Vlad, Balan Horia; Nina, Burger; Christoph, Burger-Scheidlin; Vinod, Devanathan; Dumitru, Erhan; Ioan, Hepes; Aakash, Jain; Premvir, Jain; Benjamin, Liebald; Ge, Luksys
The IUB 2002 Smallsize League Team Book Chapter
In: Kaminka, Gal; Lima, Pedro U.; Rojas, Raul (Ed.): RoboCup-02: Robot Soccer World Cup VI, Springer, 2002.
@inbook{birk_etal_f180team_rcup02,
title = {The IUB 2002 Smallsize League Team},
author = {Andreas Birk and Holger Kenn and Martijn Rooker and Agrawal Akhil and Balan Horia Vlad and Burger Nina and Burger-Scheidlin Christoph and Devanathan Vinod and Erhan Dumitru and Hepes Ioan and Jain Aakash and Jain Premvir and Liebald Benjamin and Luksys Ge},
editor = {Gal Kaminka and Pedro U. Lima and Raul Rojas},
year = {2002},
date = {2002-01-01},
booktitle = {RoboCup-02: Robot Soccer World Cup VI},
publisher = {Springer},
series = {LNAI},
keywords = {},
pubstate = {published},
tppubtype = {inbook}
}
Birk, Andreas; Kenn, Holger; Steels, Luc
Programming with Behavior Processes Journal Article
In: International Journal of Robotics and Autonomous Systems, vol. 39, pp. 115–127, 2002.
@article{behprog_ras02,
title = {Programming with Behavior Processes},
author = {Andreas Birk and Holger Kenn and Luc Steels},
url = {https://www.researchgate.net/publication/220143451_Programming_with_behavior-processes},
doi = {10.1016/S0921-8890(02)00198-7},
year = {2002},
date = {2002-01-01},
urldate = {2002-01-01},
journal = {International Journal of Robotics and Autonomous Systems},
volume = {39},
pages = {115--127},
abstract = {The so-called CubeOS is a special software environment for behavior-oriented robotics. It ranges from a dedicated nano-kernel and hardware drivers for a broad set of sensors and actuators over operating system support for concurrent and real-time programming to a special high-level language suited for novices in the field. As most special feature, the CubeOS framework includes a novel scheduler, designed for the particular needs of behavior-oriented robotics.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The so-called CubeOS is a special software environment for behavior-oriented robotics. It ranges from a dedicated nano-kernel and hardware drivers for a broad set of sensors and actuators over operating system support for concurrent and real-time programming to a special high-level language suited for novices in the field. As most special feature, the CubeOS framework includes a novel scheduler, designed for the particular needs of behavior-oriented robotics.
Birk, Andreas; Wiernik, Julie
An N-Players Prisoner's Dilemma in a Robotic Ecosystem Journal Article
In: International Journal of Robotics and Autonomous Systems, vol. 39, pp. 223–233, 2002.
@article{cnpd_ras02,
title = {An N-Players Prisoner's Dilemma in a Robotic Ecosystem},
author = {Andreas Birk and Julie Wiernik},
url = {https://www.researchgate.net/publication/222137024_An_N-Player_Prisoner's_Dilemma_in_a_Robotic_Ecosystem},
doi = {10.1016/S0921-8890(02)00206-3},
year = {2002},
date = {2002-01-01},
urldate = {2002-01-01},
journal = {International Journal of Robotics and Autonomous Systems},
volume = {39},
pages = {223--233},
abstract = {The iterated prisoner’s dilemma (iPD) in its standard form is a very well-known popular basis for research on cooperation. Here, a significant generalization of the iPD is presented in the form of an N-player game, instead of only two players, with continuous degrees of cooperation, instead of binary cooperation or defection. This continuous N-player PD (CN-PD) is motivated and explained by an actual experimental set-up in the form of an artificial ecosystem including mobile robots. Furthermore, the novel strategy of the so-called Justified-Snobism (JS) is introduced, which tries to cooperate slightly more than the average of the group of players. Results from sets of experiments with N=20 are presented that indicate that JS is indeed a successful strategy in the CN-PD.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The iterated prisoner’s dilemma (iPD) in its standard form is a very well-known popular basis for research on cooperation. Here, a significant generalization of the iPD is presented in the form of an N-player game, instead of only two players, with continuous degrees of cooperation, instead of binary cooperation or defection. This continuous N-player PD (CN-PD) is motivated and explained by an actual experimental set-up in the form of an artificial ecosystem including mobile robots. Furthermore, the novel strategy of the so-called Justified-Snobism (JS) is introduced, which tries to cooperate slightly more than the average of the group of players. Results from sets of experiments with N=20 are presented that indicate that JS is indeed a successful strategy in the CN-PD.