MPC strategy for dynamic stabilization of preplanned walking gaits

Abstract

In this paper, we report the implementation and the experimental validation of a balancing controller of bipedal robots during the execution of predefined walking patterns. The proposed controller is a cascade controller that uses the actual centre of mass (CoM) states at each sampling time and the desired CoM trajectory within a defined time window. The purpose of this controller is to generate at each sampling time a corrective term at the pelvis that allows a better tracking of the CoM and Zero Moment Point trajectories. Therefore, the overall stability is increased during the gait execution. The method permits to minimize tracking errors due to small disturbances and control errors. The effectiveness of the proposed controller is validated in simulation and in real implementation on the full-body humanoid robot Walk-Man.