Adaptive Dynamic Programming-Based Sliding Mode Optimal Position-Force Control for Reconfigurable Manipulators with Uncertain Disturbance

Abstract

In this paper, a sliding mode optimal position-force control method is proposed for constrained reconfigurable manipulators with disturbance based on adaptive dynamic programming (ADP) method and policy iteration (PI) algorithm. The sliding mode control is developed to compensate the model uncertainties, and the improved performance index function includes the sliding mode function and disturbance function. Then, the position-force control problem of the reconfigurable manipulator system under disturbance is transformed into an optimal control issue. The solution of Hamiltonian-Jacobi-Bellman (HJB) equation can be solved by using ADP and PI methods, and then the approximated sliding mode optimal control policy can be derived by constructing the critic neural network (NN). The closed-loop robotic system is proved to be asymptotic stability by using Lyapunov theory. Finally, simulations are provided to demonstrate the effectiveness of the proposed method.