Reinforcement Learning Based, Staircase Negotiation Learning: Simulation and Transfer to Reality for Articulated Tracked Robots

Abstract

Autonomous control of reconfigurable robots is crucial for their deployment in diverse environments. However, its development is hampered by the diversity of hardware and task constraints. We advocate the use of artificial intelligence-based approaches to improve scalability to different conditions and portability to platforms of comparable traversability skills. In particular, we succeed in tackling the problem of staircase traversal via a reinforcement learning (RL)-based control framework applicable to different articulated, tracked robots and powerful enough to generalize to varying conditions learned in simulation and transferred to reality in a zero-shot setting. Our extensive experiments demonstrate the robustness of the framework in learning tasks with increased risk and difficulty induced by platform diversification and increased control dimensionality.