Abstract
Foothold identification is a key ability for legged robots that allows generating terrain adaptive behaviors (e.g., gait and control parameters) and thereby improving mobility in complex environment. To this end, this paper addresses the issue of foothold characterization and identification over rugged terrain, from the terrain geometry point of view. For a terrain region that might be a potential foothold of a robotic leg, the characteristic features are extracted as two first-order partial derivatives and two curvature parameters of a quadric regression surface at this location. These features are able to give an intuitive and, more importantly, accurate characterization towards the specific geometry of the ground location. On this basis, a supervised learning technique, Support Vector Machine (SVM), is employed, seeking to learn a foothold identification policy from human expert demonstration. As a result, an SVM classifier is learnt using the extracted features and human-demonstrated labels, which is able to identify whether or not a certain ground location is suited as a safe foot support for a robotic leg. It is shown that over 90% identification rate can be achieved with the proposed approach. Finally, preliminary experiment is implemented with a six-legged robot to demonstrate the effectiveness of the proposed approach.
Similar content being viewed by others
References
Yu B, Ba K X, Liu Y L, Jin Z G, Zhu Q X, Gao Z J, Ma G L, Kong X D. Design and application of MVIC for hydraulic drive unit of legged robot. IEEE Access, 2019, 7, 11237–11245.
Hyun D J, Lee J, Park S, Kim S. Implementation of trot-to-gallop transition and subsequent gallop on the MIT Cheetah I. The International Journal of Robotics Research, 2016, 35, 1627–1650.
Chen J, Jin H Z, Iida F, Zhao J. A design concept of parallel elasticity extracted from biological muscles for engineered actuators. Bioinspiration & Biomimetics, 2016, 11, 056009.
Bartsch S, Birnschein T, Römmermann M, Hilljegerdes J, Kühn D, Kirchner F. Development of the six-legged walking and climbing robot SpaceClimber. Journal of Field Robotics, 2012, 29, 506–532.
Zhang J Q, Gao F, Han X L, Chen X B, Han X Y. Trot gait design and CPG method for a quadruped robot. Journal of Bionic Engineering, 2014, 11, 18–25.
Chen J, Liang Z C, Zhu Y H, Liu C, Zhang L, Hao L N, Zhao J. Towards the exploitation of physical compliance in segmented and electrically actuated robotic legs: A review focused on elastic mechanisms. Sensors, 2019, 19, 5351.
Chen J, Liu Y B, Liu G F, Zhao J, Liu Y B, Liu G F. On the utility of leg distal compliance for buffering landing impact of legged robots. Advances in Mechanical Engineering, 2017, 9, 1687814017700058.
Stelzer A, Hirschmüller H, Görner M. Stereo-vision-based navigation of a six-legged walking robot in unknown rough terrain. The International Journal of Robotics Research, 2012, 31, 381–402.
Belter D. Efficient modeling and evaluation of constraints in path planning for multi-legged walking robots. IEEE Access, 2019, 7, 107845–107862.
Belter D, Skrzypczynski P. Rough terrain mapping and classification for foothold selection in a walking robot. Journal of Field Robotics, 2011, 28, 497–528.
Chen J, Gao F, Huang C, Zhao J. Whole-body motion planning for a six-legged robot walking on rugged terrain. Applied Sciences, 2019, 9, 5284.
Wellhausen L, Dosovitskiy A, Ranftl R, Walas K, Cadena C, Hutter M. Where should I walk? Predicting terrain properties from images via self-supervised learning. IEEE Robotics and Automation Letters, 2019, 4, 1509–1516.
Prágr M, Čížek P, Faigl J. Cost of transport estimation for legged robot based on terrain features inference from aerial scan. Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, Madrid, Spain, 2018, 1745–1750.
Prágr M, Čížek P, Bayer J, Faigl J. Online incremental learning of the terrain traversal cost in autonomous exploration. Robotics: Science and Systems XV, Freiburg im Breisgau, Germany, 2019.
Kolter J Z, Ng A Y. The Stanford LittleDog: A learning and rapid replanning approach to quadruped locomotion. The International Journal of Robotics Research, 2011, 30, 150–174.
Kalakrishnan M, Buchli J, Pastor P, Mistry M, Schaal S. Learning, planning, and control for quadruped locomotion over challenging terrain. The International Journal of Robotics Research, 2010, 30, 236–258.
Murphy M P, Saunders A, Moreira C, Rizzi A A, Raibert M. The littledog robot. The International Journal of Robotics Research, 2010, 30, 145–149.
Hoepflinger M A, Remy C D, Hutter M, Spinello L, Siegwart R. Haptic terrain classification for legged robots. Proceedings of IEEE International Conference on Robotics and Automation, Anchorage, USA, 2010, 2828–2833.
Hoepflinger M A, Hutter M, Gehring C, Bloesch M, Siegwart R. Unsupervised identification and prediction of foothold robustness. Proceedings of IEEE International Conference on Robotics and Automation, Karlsruhe, Germany, 2013, 3293–3298.
Zhang H, Liu Y B, Zhao J, Chen J, Yan J H. Development of a bionic hexapod robot for walking on unstructured terrain. Journal of Bionic Engineering, 2014, 11, 176–187.
Chen J, Liu Y B, Zhao J, Zhang H, Jin H Z. Biomimetic design and optimal swing of a hexapod robot leg. Journal of Bionic Engineering, 2014, 11, 26–35.
Chen J, Liang Z C, Zhu Y H, Zhao J. Improving kinematic flexibility and walking performance of a six-legged robot by rationally designing leg morphology. Journal of Bionic Engineering, 2019, 16, 608–620.
Zhang H, Wu R, Li C L, Zang X Z, Zhang X H, Jin H Z, Zhao J. A force-sensing system on legs for biomimetic hexapod robots interacting with unstructured terrain. Sensors, 2017, 17, 1514.
Vapnik V. The Nature of Statistical Learning Theory, Springer Science & Business Media, New York, USA, 2013.
Burges C J. A tutorial on support vector machines for pattern recognition. Data Mining and Knowledge Discovery, 1998, 2, 121–167.
Chang C C, Lin C J. LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2011, 2, 27.
Hutter M, Gehring C, Lauber A, Gunther F, Bellicoso C D, Tsounis V, Fankhauser P, Diethelm R, Bachmann S, Blösch M, Kolvenbach H. ANYmal — Toward legged robots for harsh environments. Advanced Robotics, 2017, 31, 918–931.
Hutter M, Gehring C, Jud D, Lauber A, Bellicoso C D, Tsounis V, Hwangbo J, Bodie K, Fankhauser P, Bloesch M, Diethelm R. ANYmal — A highly mobile and dynamic quadrupedal robot. Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems, Daejeon, South Korea, 2016, 38–44.
Bednarek J, Bednarek M, Wellhausen L, Hutter M, Walas K. What am I touching? Learning to classify terrain via haptic sensing. Proceedings of International Conference on Robotics and Automation, Montreal, Canada, 2019, 7187–7193.
Belter D, Łabȩcki P, Skrzypczyński P. Adaptive motion planning for autonomous rough terrain traversal with a walking robot. Journal of Field Robotics, 2016, 33, 337–370.
Tian Y, Gao F. Efficient motion generation for a six-legged robot walking on irregular terrain via integrated foothold selection and optimization-based whole-body planning. Robotica, 2018, 36, 333–352.
Xi W, Yesilevskiy Y, Remy C D. Selecting gaits for economical locomotion of legged robots. The International Journal of Robotics Research, 2016, 35, 1140–1154.
Hoyt D F, Taylor C R. Gait and the energetics of locomotion in horses. Nature, 1984, 292, 239–240.
Cruse H. What mechanisms coordinate leg movement in walking arthropods? Trends in Neurosciences, 1990, 13, 15–21.
Cruse H, Kindermann T, Schumm M, Dean J, Schmitz J. Walknet — A biologically inspired network to control six-legged walking. Neural Networks, 1998, 11, 1435–1447.
Acknowledgment
This work was supported by National Natural Science Foundation of China (Grant No. 51805074), State Key Laboratory of Robotics and System (HIT) (Grant No. SKLRS-2018-KF-02), China postdoctoral Science Foundation (Grant Nos. 2018M631799 and 2019T120213), Fundamental Research Funds for the Central Universities (Grant No. N2003001), Natural Science Foundation of Liaoning Province (Grant No. 2019-BS-090).
Author information
Authors and Affiliations
Corresponding authors
Rights and permissions
About this article
Cite this article
Chen, J., Liu, C., Zhao, H. et al. Learning to Identify Footholds from Geometric Characteristics for a Six-legged Robot over Rugged Terrain. J Bionic Eng 17, 512–522 (2020). https://doi.org/10.1007/s42235-020-0041-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42235-020-0041-4