A Simple Rebalance Strategy for Omnidirectional Humanoids Walking by Learning Foot Positioning

Xu Tao,Chen Qijun
2011-01-01
Abstract:On solving the rebalance problem of the trajectory-based humanoids walking approaches, a simple foot positioning compensator is proposed to modify the foot positioning online based on the estimated robot state using onboard sensors. To make the compensator coincident with the dynamics of a full-body humanoid robot, the foot positioning policy is learnt through a policy gradient reinforcement learning approach. Experiments on both simulated and real full-body humanoid robots validate the good performance of the proposed method not only in forward walking but also in omnidirectional walking.
What problem does this paper attempt to address?