Learning footstep planning on irregular surfaces with partial placements

We present two contributions built upon on a previous footstep planner based on the ARA* search. Firstly, we have developed an improved foothold selection method using support polygons, to increase foothold availability in rough terrain. Secondly, we present a footstep classification method using the C5.0 algorithm, that takes advantage of cost similarity between adjacent steps. This is intended to learn feasibility and approximate transition costs for the ARA* planner.These contributions extend capabilities of the planner by increasing footstep availability and allowing to generate more complex plans, without compromising safety.

[1]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[2]  Oskar von Stryk,et al.  Supervised footstep planning for humanoid robots in rough terrain tasks using a black box walking controller , 2014, 2014 IEEE-RAS International Conference on Humanoid Robots.

[3]  Andrew Y. Ng,et al.  A control architecture for quadruped locomotion over rough terrain , 2008, 2008 IEEE International Conference on Robotics and Automation.

[4]  Maren Bennewitz,et al.  Anytime search-based footstep planning with suboptimality bounds , 2012, 2012 12th IEEE-RAS International Conference on Humanoid Robots (Humanoids 2012).

[5]  Sylvain Bertrand,et al.  Walking on partial footholds including line contacts with the humanoid robot atlas , 2016, 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids).

[6]  Piotr Skrzypczynski,et al.  Adaptive Motion Planning for Autonomous Rough Terrain Traversal with a Walking Robot , 2016, J. Field Robotics.

[7]  Piotr Skrzypczynski,et al.  Rough terrain mapping and classification for foothold selection in a walking robot , 2010, 2010 IEEE Safety Security and Rescue Robotics.

[8]  Johannes Garimort,et al.  Humanoid navigation with dynamic footstep plans , 2011, 2011 IEEE International Conference on Robotics and Automation.

[9]  Christopher G. Atkeson,et al.  Optimization and learning for rough terrain legged locomotion , 2011, Int. J. Robotics Res..

[10]  Steven M. LaValle,et al.  Randomized Kinodynamic Planning , 1999, Proceedings 1999 IEEE International Conference on Robotics and Automation (Cat. No.99CH36288C).

[11]  Maxim Likhachev,et al.  D*lite , 2002, AAAI/IAAI.

[12]  Oussama Khatib,et al.  Contact-consistent elastic strips for multi-contact locomotion planning of humanoid robots , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[13]  Jaime Valls Miró,et al.  Planning Stable Paths for Urban Search and Rescue Robots , 2012, RoboCup.

[14]  Sebastian Thrun,et al.  ARA*: Anytime A* with Provable Bounds on Sub-Optimality , 2003, NIPS.

[15]  Daniel Maier,et al.  Integrated perception, mapping, and footstep planning for humanoid navigation among 3D obstacles , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[16]  Evangelos Papadopoulos,et al.  A new measure of tipover stability margin for mobile manipulators , 1996, Proceedings of IEEE International Conference on Robotics and Automation.

[17]  Nikolaos G. Tsagarakis,et al.  Footstep Planning in Rough Terrain for Bipedal Robots Using Curved Contact Patches , 2018, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[18]  Oskar von Stryk,et al.  Open source integrated 3D footstep planning framework for humanoid robots , 2016, 2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids).