Impedance and Trajectory Adaptation for Contact Robots Using Integral Reinforcement Learning

Abstract

In this paper, we develop a learning controller that adapts and tracks the impedance and trajectory for robots interacting with unknown environments. Impedance adaptation is used to compensate for contacting with the environment, while the reference trajectory learning is to maintain a prescribed interaction force. The tracking performance is ensured by an adaptive learning controller with Integral Reinforcement learning (IRL) for partially unknown system dynamics. The contact dynamics are analysed via Lyapunov theory and the effectiveness of the proposed control method is verified through simulations.