Optimized Adaptive Self-Triggered Fault-Tolerant Control for Markov Jumping Nonlinear Systems via Reinforcement Learning and Its Application to Tunnel Diode Circuits

Abstract

In this paper, the optimized adaptive fault-tolerant self-triggered control design for Markov jumping nonlinear systems with assured performance has been studied via reinforcement learning (RL). Initially, by means of the convergence properties of the utilized prescribed performance function, the tracking error is driven into the expected steady-state interval at a fixed time with the specific evolving behavior. Additionally, RL is introduced into recursive design procedure to achieve a potential balance between performance improvement and control costs based on the actor-critic architecture. Furthermore, combining with the self-triggered mechanism and fault-tolerant mechanism, an optimized adaptive resilient controller with self-triggered characteristics is designed to ensure that all the signals in the closed-loop system are bounded in probability, and the tracking error can be regulated into a predetermined region satisfying the pre-specified tracking accuracy within a fixed time even if the actuator faults occur suddenly. Eventually, two illustrative examples including a numerical model and a practical tunnel diode circuit model are adopted to demonstrate the validity of the designed scheme.