Cooperative fault-tolerant tracking control for multi-agent systems: A multiple description encoding scheme

• Published in: Applied mathematics and computation Vol. 462; p. 128337
• Main Authors: Wang, Xi, Ju, Yamei, Ding, Derui, Liu, Hongjian
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.02.2024
• Subjects: Cooperative control; Fault-tolerant tracking control; Intermediate estimators; Multi-agent systems; Multiple description encoding schemes; Intermediate estimators; Fault-tolerant tracking control; Cooperative control; Multiple description encoding schemes; Multi-agent systems
• ISSN: 0096-3003; 1873-5649
• DOI: 10.1016/j.amc.2023.128337

In this article, the cooperative fault-tolerant tracking control (FTTC) is investigated for discrete-time multi-agent systems (MASs) with time-varying delays (TVDs) under multiple description encoding schemes (MDESs). First, a uniform channel model is proposed to describe the employed MDES subject to the effect of packet dropouts by introducing two independent random variables obeying the Bernoulli distribution and three indicator variables. Subsequently, a novel intermediate estimator is designed to estimate both system states and a fictitious intermediate variable (an integration of faults and leader's inputs) based on relatively measured outputs. In terms of the Lyapunov stability theory, some sufficient conditions are acquired to guarantee that the closed-loop system is exponentially ultimately bounded in the mean-square sense. Furthermore, the desired gain matrices are obtained by resorting to both the graph feature and singular value decomposition. Finally, the effectiveness and superiority are tested by two simulation examples for the proposed tracking protocol. •An intermediate estimator by introducing intermediate variables is designed to estimate system states and all unknown input signals.•Some sufficient conditions are obtained such that tracking errors of MASs with MDESs achieve exponentially ultimate boundedness.•The desired gain matrices are obtained by resorting to the graph feature and singular value decomposition.