Sliding Mode Control Under Redundant Channels: Handling Markov Packet Dropouts

Abstract

This article is concerned with the sliding mode control (SMC) problem for a class of Markov jump systems subject to packet dropouts, in which the dropped or received status of packet is described by a Markov chain. To enhance the reliability of data transmission, multiple redundant channels are employed between sensors and the controller. Different from the existing measurement model under redundant channels with Bernoulli-process-based packet dropouts, a novel measurement model under Markov-chain-based packet dropouts is proposed under the redundant channel transmission. It is assumed that the modes of the controlled system and packet-dropout model are unavailable, and then a mode detection mechanism is proposed to detect the partially unavailable modes. By utilizing the detected modes, a dynamic observer is constructed to estimate the unmeasurable system state, based on which a detected-mode-dependent sliding mode controller is designed to achieve the mean-square exponential ultimate boundedness of the closed-loop system. Meanwhile, incremental search technique and particle swarm optimization algorithm are, respectively, utilized to solve two optimization problems for enhancing the closed-loop performances. Finally, two simulation examples are provided to verify the effectiveness of the proposed schemes.