Adaptive Impedance Control of Human–Robot Cooperation Using Reinforcement Learning

摘要

This paper presents human–robot cooperation with adaptive behavior of the robot, which helps the human operator to perform the cooperative task and optimizes its performance. A novel adaptive impedance control is proposed for the robotic manipulator, whose end-effector's motions are constrained by human arm motion limits. In order to minimized motion tracking errors and acquire an optimal impedance mode of human arms, the linear quadratic regulation (LQR) is formulated; then, integral reinforcement learning (IRL) has been proposed to solve the given LQR with little information of the human arm model. Considering human–robot interaction force during the robot performing manipulation, a novel barrier-Lyapunov-function-based adaptive impedance control incorporating adaptive parameter learning is developed for physical limits, transient perturbations, and time-varying dynamics. Experimental results validate that the proposed controller is effective in assisting the operator to perform the human–robot cooperative task.