Design of Nonlinear Iterative Learning Control Based on Deep Reinforcement Learning Algorithm

Abstract

Iterative learning control (ILC) is an advanced control method which has been studied and widely used in periodical, repetitive or batch processes. However, there is still a lack of an effective method for the design of nonlinear iterative learning control (NILC). In view of the excellent performance of deep reinforcement learning (DRL) in dealing with the decision-making problems for the complex dynamical processes, in this paper, we propose an intelligent design method for NILC system by using deep deterministic policy gradient (DDPG), a typical DRL algorithm. By properly designing the state information and the instant reward, the design algorithm can gradually realize the optimal NILC law through the interaction without any requirement of prior knowledge of the processes. The numerical simulation based on a nonlinear model illustrates the effectiveness and applicability of the proposed design method.