Finite-Time Active Disturbance Rejection Trajectory Tracking Control of Robot Manipulators via PPO Algorithm

This paper studies the trajectory tracking control problem of robot manipulator by using reinforcement learning algorithms. First, for the case of robot system without disturbance, a finite-time trajectory tracking controller is designed based on homogeneous system theory. Then, in the presence of uncertain disturbances, an uncertainty-robust algorithm based on PPO (Proximal Policy Optimization) is developed. Based on the previous results, a finite-time disturbance-robust trajectory tracking controller based on PPO is proposed. A rigorous analysis proves the stability of the closed-loop system. Finally, simulation results verify the effectiveness of the theoretical analysis.