\n \n \n
\n
\n\n \n \n \n \n \n \n Advances in Real-World Applications for Legged Robots.\n \n \n \n \n\n\n \n Bellicoso, C. D.; Bjelonic, M.; Wellhausen, L.; Sako, D.; Holtmann, K.; Guenther, F.; Tranzatto, M.; Fankhauser, P.; and Hutter, M.\n\n\n \n\n\n\n
Journal of Field Robotics, 35(8): 1311-1326. 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n
pdf\n \n \n \n
video\n \n \n \n
link\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 3 downloads\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@article{bellicoso2018advances,\n author = {Bellicoso, C. Dario and\n Bjelonic, Marko and\n Wellhausen, Lorenz and\n Sako, Dhionis and\n Holtmann, Kai and\n Guenther, Fabian and\n Tranzatto, Marco and\n Fankhauser, P{\\'e}ter and\n Hutter, Marco},\n title = {Advances in Real-World Applications for Legged Robots},\n journal = {Journal of Field Robotics},\n volume = {35},\n number = {8},\n pages = {1311-1326},\n doi = {10.1002/rob.21839},\n year = {2018},\n abstract = {This paper provides insight into the application of the\n quadrupedal robot ANYmal in outdoor missions of industrial\n inspection (ARGOS Challenge) and search and rescue (European\n Robotics League (ERL) Emergency Robots). In both competitions,\n the legged robot had to autonomously and semi-autonomously\n navigate in real-world scenarios to complete high-level tasks\n such as inspection and payload delivery. In the ARGOS\n competition, ANYmal used a rotating LiDAR sensor to\n localize on the industrial site and map the terrain and obstacles\n around the robot. In the ERL competition, additional Real-Time\n Kinematic (RTK)-Global Positioning System (GPS) was used to\n co-localize the legged robot with respect to a Micro Aerial\n Vehicle (MAV) that creates maps from the aerial view. The high\n mobility of legged robots allows overcoming large obstacles, e.g.\n steps and stairs, with statically and dynamically stable gaits.\n Moreover, the versatile machine can adapt its posture for\n inspection and payload delivery. The paper concludes with insight\n into the general learnings from the ARGOS and ERL challenges.},\n keywords = {legged robot, quadrupedal robot, localization, mapping,\n challenge},\n url_pdf = {files/bellicoso2018advances.pdf},\n url_video = {https://youtu.be/qrJlMze_xhQ},\n url_link = {https://doi.org/10.1002/rob.21839}\n}\n\n
\n
\n\n\n
\n This paper provides insight into the application of the quadrupedal robot ANYmal in outdoor missions of industrial inspection (ARGOS Challenge) and search and rescue (European Robotics League (ERL) Emergency Robots). In both competitions, the legged robot had to autonomously and semi-autonomously navigate in real-world scenarios to complete high-level tasks such as inspection and payload delivery. In the ARGOS competition, ANYmal used a rotating LiDAR sensor to localize on the industrial site and map the terrain and obstacles around the robot. In the ERL competition, additional Real-Time Kinematic (RTK)-Global Positioning System (GPS) was used to co-localize the legged robot with respect to a Micro Aerial Vehicle (MAV) that creates maps from the aerial view. The high mobility of legged robots allows overcoming large obstacles, e.g. steps and stairs, with statically and dynamically stable gaits. Moreover, the versatile machine can adapt its posture for inspection and payload delivery. The paper concludes with insight into the general learnings from the ARGOS and ERL challenges.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Real-Time Dance Generation to Music for a Legged Robot.\n \n \n \n \n\n\n \n Bi, T.; Fankhauser, P.; Bellicoso, C. D.; and Hutter, M.\n\n\n \n\n\n\n In
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1038-1044, oct 2018. \n
\n\n
\n\n
\n\n
\n\n \n \n
video\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n\n\n\n
\n
@inproceedings{bi2018realtime, \nauthor = {Bi, Thomas and\n Fankhauser, P{\\'e}ter and\n Bellicoso, C. Dario and\n Hutter, Marco}, \nbooktitle = {2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, \ntitle = {Real-Time Dance Generation to Music for a Legged Robot}, \nyear = {2018}, \nvolume = {}, \nnumber = {}, \npages = {1038-1044}, \nkeywords = {feedback;feedforward;humanoid robots;image motion analysis;legged locomotion;Markov processes;motion control;music;robot vision;synchronisation;music tempo;dance generation;feedforward delay controller;Markov chain;quadrupedal robot;robot whole-body controller reference input;feedback delay controller;time-shifting;delays;picked dance motion;base motions;stepping motions;dance motions;user-generated dance motion library;dance choreography;onboard microphone;live music;external stimuli;legged robot;Robot kinematics;Legged locomotion;Music;Delays;Trajectory;Real-time systems}, \ndoi = {10.1109/IROS.2018.8593983}, \nISSN = {2153-0866}, \nmonth = {oct},\nurl_video = {https://youtu.be/kHBLaw5nfzk},\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Skating with a force controlled quadrupedal robot.\n \n \n \n \n\n\n \n Bjelonic, M.; Bellicoso, C. D.; Tiryaki, M E.; and Hutter, M.\n\n\n \n\n\n\n In
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 7555–7561, 2018. \n
\n\n
\n\n
\n\n
\n\n \n \n
pdf\n \n \n \n
video\n \n \n \n
link\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 6 downloads\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{bjelonic2018skating,\n author = {Bjelonic, Marko and\n Bellicoso, C. Dario and\n Tiryaki, M Efe and\n Hutter, Marco},\n title = {Skating with a force controlled quadrupedal robot},\n booktitle = {IEEE/RSJ International Conference on Intelligent Robots and\n Systems (IROS)},\n year = {2018},\n pages = {7555--7561},\n doi = {10.1109/IROS.2018.8594504},\n abstract = {Traditional legged robots are capable of traversing challenging\n terrain, but lack of energy efficiency when compared to wheeled\n systems operating on flat environments. The combination of both\n locomotion domains overcomes the trade-off between mobility and\n efficiency. Therefore, this paper presents a novel motion planner\n and controller which together enable a legged robot equipped with\n skates to perform skating maneuvers. These are achieved by an\n appropriate combination of planned reaction forces and gliding\n motions. Our novel motion controller formulates a Virtual Model\n Controller and an optimal contact force distribution which takes\n into account the nonholonomic constraints introduced by the\n skates. This approach has been tested on the torque-controllable\n robot ANYmal equipped with passive wheels and ice skates as\n end-effectors. We conducted experiments on flat and inclined\n terrain, whereby we show that skating motions reduces the cost\n of transport by up to 80{\\,\\%} with respect to traditional walking\n gaits.},\n keywords = {legged locomotion, quadrupedal robot, skating, motion control,\n force control},\n url_pdf = {files/bjelonic2018skating.pdf},\n url_video = {https://youtu.be/fJfAWiylpxw},\n url_link = {https://ieeexplore.ieee.org/document/8594504}\n}\n\n
\n
\n\n\n
\n Traditional legged robots are capable of traversing challenging terrain, but lack of energy efficiency when compared to wheeled systems operating on flat environments. The combination of both locomotion domains overcomes the trade-off between mobility and efficiency. Therefore, this paper presents a novel motion planner and controller which together enable a legged robot equipped with skates to perform skating maneuvers. These are achieved by an appropriate combination of planned reaction forces and gliding motions. Our novel motion controller formulates a Virtual Model Controller and an optimal contact force distribution which takes into account the nonholonomic constraints introduced by the skates. This approach has been tested on the torque-controllable robot ANYmal equipped with passive wheels and ice skates as end-effectors. We conducted experiments on flat and inclined terrain, whereby we show that skating motions reduces the cost of transport by up to 80\\,% with respect to traditional walking gaits.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n Efficient Gait Selection for Quadrupedal Robots on the Moon and Mars.\n \n \n \n\n\n \n Kolvenbach, H.; Bellicoso, C. D.; Jenelten, F.; Wellhausen, L.; and Hutter, M.\n\n\n \n\n\n\n In
14th International Symposium on Artificial Intelligence, Robotics and Automation in Space (i-SAIRAS 2018), 2018. ESA Conference Bureau\n
\n\n
\n\n
\n\n
\n\n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@inproceedings{kolvenbach2018efficient,\n title={Efficient Gait Selection for Quadrupedal Robots on the Moon and Mars},\n author={Kolvenbach, Hendrik and Bellicoso, C. Dario and Jenelten, Fabian and Wellhausen, Lorenz and Hutter, Marco},\n booktitle={14th International Symposium on Artificial Intelligence, Robotics and Automation in Space (i-SAIRAS 2018)},\n year={2018},\n organization={ESA Conference Bureau}\n}\n\n
\n
\n\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Robust Rough-Terrain Locomotion with a Quadrupedal Robot.\n \n \n \n \n\n\n \n Fankhauser, P.; Bjelonic, M.; Bellicoso, C. D.; Miki, T.; and Hutter, M.\n\n\n \n\n\n\n In
IEEE International Conference on Robotics and Automation (ICRA), 2018. \n
\n\n
\n\n
\n\n
\n\n \n \n
video\n \n \n \n
link\n \n \n\n \n \n doi\n \n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n \n \n 4 downloads\n \n \n\n \n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n
@inproceedings{fankhauser2018robust,\n author = {Fankhauser, P{\\'e}ter and\n Bjelonic, Marko and\n Bellicoso, C. Dario and\n Miki, Takahiro and\n Hutter, Marco},\n title = {Robust Rough-Terrain Locomotion with a Quadrupedal Robot},\n booktitle = {IEEE International Conference on Robotics and Automation\n (ICRA)},\n year = {2018},\n doi = {10.1109/ICRA.2018.8460731},\n abstract = {Robots working in natural, urban, and industrial settings need to\n be able to navigate challenging environments. In this paper, we\n present a motion planner for the perceptive rough-terrain\n locomotion with quadrupedal robots. The planner finds safe\n footholds along with collision-free swing-leg motions by\n leveraging an acquired terrain map. To this end, we present a\n novel pose optimization approach that enables the robot to climb\n over significant obstacles. We experimentally validate our\n approach with the quadrupedal robot ANYmal by autonomously\n traversing obstacles such steps, inclines, and stairs. The\n locomotion planner re-plans the motion at every step to cope with\n disturbances and dynamic environments. The robot has no prior\n knowledge of the scene, and all mapping, state estimation,\n control, and planning is performed in real-time onboard the\n robot.},\n keywords = {legged locomotion, robot sensing systems, planning,\n collision avoidance},\n url_video = {https://youtu.be/CpzQu25iLa0},\n url_link = {https://ieeexplore.ieee.org/document/8460731},\n}\n\n
\n
\n\n\n
\n Robots working in natural, urban, and industrial settings need to be able to navigate challenging environments. In this paper, we present a motion planner for the perceptive rough-terrain locomotion with quadrupedal robots. The planner finds safe footholds along with collision-free swing-leg motions by leveraging an acquired terrain map. To this end, we present a novel pose optimization approach that enables the robot to climb over significant obstacles. We experimentally validate our approach with the quadrupedal robot ANYmal by autonomously traversing obstacles such steps, inclines, and stairs. The locomotion planner re-plans the motion at every step to cope with disturbances and dynamic environments. The robot has no prior knowledge of the scene, and all mapping, state estimation, control, and planning is performed in real-time onboard the robot.\n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Gait and trajectory optimization for legged systems through phase-based end-effector parameterization.\n \n \n \n \n\n\n \n Winkler, A. W; Bellicoso, C. D.; Hutter, M.; and Buchli, J.\n\n\n \n\n\n\n
IEEE Robotics and Automation Letters, 3(3): 1560–1567. 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n
video\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{winkler2018gait,\n title = {Gait and trajectory optimization for legged systems through phase-based end-effector parameterization},\n abstract = {\n We present a single trajectory optimization formulation for legged locomotion that automatically determines the gait sequence, step timings, footholds, swing-leg motions, and six-dimensional body motion over nonflat terrain, without any additional modules. Our phase-based parameterization of feet motion and forces allows to optimize over the discrete gait sequence using only continuous decision variables. The system is represented using a simplified centroidal dynamics model that is influenced by the feet's location and forces. We explicitly enforce friction cone constraints, depending on the shape of the terrain. The nonlinear programming problem solver generates highly dynamic motion plans with full flight phases for a variety of legged systems with arbitrary morphologies in an efficient manner. We validate the feasibility of the generated plans in simulation and on the real quadruped robot ANYmal. Additionally, the entire solver software TOWR, which used to generate these motions is made freely available.\n },\n author = {Winkler, Alexander W and Bellicoso, C. Dario and Hutter, Marco and Buchli, Jonas},\n journal = {IEEE Robotics and Automation Letters},\n volume = {3},\n number = {3},\n pages = {1560--1567},\n year = {2018},\n publisher = {IEEE},\n url_video = {https://youtu.be/0jE46GqzxMM},\n}\n\n
\n
\n\n\n
\n We present a single trajectory optimization formulation for legged locomotion that automatically determines the gait sequence, step timings, footholds, swing-leg motions, and six-dimensional body motion over nonflat terrain, without any additional modules. Our phase-based parameterization of feet motion and forces allows to optimize over the discrete gait sequence using only continuous decision variables. The system is represented using a simplified centroidal dynamics model that is influenced by the feet's location and forces. We explicitly enforce friction cone constraints, depending on the shape of the terrain. The nonlinear programming problem solver generates highly dynamic motion plans with full flight phases for a variety of legged systems with arbitrary morphologies in an efficient manner. We validate the feasibility of the generated plans in simulation and on the real quadruped robot ANYmal. Additionally, the entire solver software TOWR, which used to generate these motions is made freely available. \n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Whole-body nonlinear model predictive control through contacts for quadrupeds.\n \n \n \n \n\n\n \n Neunert, M.; Stäuble, M.; Giftthaler, M.; Bellicoso, C. D.; Carius, J.; Gehring, C.; Hutter, M.; and Buchli, J.\n\n\n \n\n\n\n
IEEE Robotics and Automation Letters, 3(3): 1458–1465. jan 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n
video\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{neunert2018whole,\n title = {Whole-body nonlinear model predictive control through contacts for quadrupeds},\n author = {Neunert, Michael and St{\\"a}uble, Markus and Giftthaler, Markus and Bellicoso, C. Dario and Carius, Jan and Gehring, Christian and Hutter, Marco and Buchli, Jonas},\n abstract = {\n In this letter, we present a whole-body nonlinear model predictive control approach for rigid body systems subject to contacts. We use a full-dynamic system model which also includes explicit contact dynamics. Therefore, contact locations, sequences, and timings are not prespecified but optimized by the solver. Yet, using numerical and software engineering allows for running the nonlinear Optimal Control solver at rates up to 190 Hz on a quadruped for a time horizon of half a second. This outperforms the state-of-the-art by at least one order of magnitude. Hardware experiments in the form of periodic and nonperiodic tasks are applied to two quadrupeds with different actuation systems. The obtained results underline the performance, transferability, and robustness of the approach.\n },\n journal = {IEEE Robotics and Automation Letters},\n volume = {3},\n number = {3},\n pages = {1458--1465},\n year = {2018},\n month = {jan},\n publisher = {IEEE},\n url_video = {https://youtu.be/mRC0x7KOnNE},\n}\n\n
\n
\n\n\n
\n In this letter, we present a whole-body nonlinear model predictive control approach for rigid body systems subject to contacts. We use a full-dynamic system model which also includes explicit contact dynamics. Therefore, contact locations, sequences, and timings are not prespecified but optimized by the solver. Yet, using numerical and software engineering allows for running the nonlinear Optimal Control solver at rates up to 190 Hz on a quadruped for a time horizon of half a second. This outperforms the state-of-the-art by at least one order of magnitude. Hardware experiments in the form of periodic and nonperiodic tasks are applied to two quadrupeds with different actuation systems. The obtained results underline the performance, transferability, and robustness of the approach. \n
\n\n\n
\n\n\n
\n
\n\n \n \n \n \n \n \n Dynamic Locomotion Through Online Nonlinear Motion Optimization for Quadrupedal Robots.\n \n \n \n \n\n\n \n Bellicoso, C. D.; Jenelten, F.; Gehring, C.; and Hutter, M.\n\n\n \n\n\n\n
IEEE Robotics and Automation Letters, 3(3): 2261–2268. jan 2018.\n
\n\n
\n\n
\n\n
\n\n \n \n
video\n \n \n\n \n\n \n link\n \n \n\n bibtex\n \n\n \n \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n \n \n \n\n\n\n
\n
@article{bellicoso2018dynamic,\n title = {Dynamic Locomotion Through Online Nonlinear Motion Optimization for Quadrupedal Robots},\n author = {Bellicoso, C. Dario and Jenelten, Fabian and Gehring, Christian and Hutter, Marco},\n abstract = {\n This letter presents a realtime motion planning and control method that enables a quadrupedal robot to execute dynamic gaits including trot, pace, and dynamic lateral walk, as well as gaits with full flight phases such as jumping, pronking, and running trot. The proposed method also enables smooth transitions between these gaits. Our approach relies on an online zero-moment point based motion planner which continuously updates the reference motion trajectory as a function of the contact schedule and the state of the robot. The reference footholds for each leg are obtained by solving a separate optimization problem. The resulting optimized motion plans are tracked by a hierarchical whole-body controller. Our framework has been tested in simulation and on ANYmal, a fully torque-controllable quadrupedal robot, both in simulation and on the actual robot.\n },\n journal = {IEEE Robotics and Automation Letters},\n volume = {3},\n number = {3},\n pages = {2261--2268},\n year = {2018},\n month = {jan},\n publisher = {IEEE},\n url_video = {https://youtu.be/wQpLzoEzsx8},\n}\n\n
\n
\n\n\n
\n This letter presents a realtime motion planning and control method that enables a quadrupedal robot to execute dynamic gaits including trot, pace, and dynamic lateral walk, as well as gaits with full flight phases such as jumping, pronking, and running trot. The proposed method also enables smooth transitions between these gaits. Our approach relies on an online zero-moment point based motion planner which continuously updates the reference motion trajectory as a function of the contact schedule and the state of the robot. The reference footholds for each leg are obtained by solving a separate optimization problem. The resulting optimized motion plans are tracked by a hierarchical whole-body controller. Our framework has been tested in simulation and on ANYmal, a fully torque-controllable quadrupedal robot, both in simulation and on the actual robot. \n
\n\n\n
\n\n\n\n\n\n