Asymptotic consensus of hybrid multi-agent systems considering network attacks

Abstract

Abstract This paper focuses on the problem of active defense and achieving asymptotic consensus control in hybrid multi-agent systems under network attacks. Considering the influence of Byzantine nodes, on the basis of the theory of mimic defense, a deep dynamic heterogeneous redundancy architecture for hybrid multi-agent systems is established. Drawing on the Lyapunov asymptotic stability theory, a hybrid multi-agent asymptotic consensus is proposed. Furthermore, a hybrid multi-agent progressive consensus clustering review strategy is suggested. Via Lyapunov asymptotic stability theory, the sufficient and necessary conditions for achieving the asymptotic consensus of hybrid multi-agent systems on the basis of a hybrid clustering review strategy are proven. The strategy’s extended application method in asymptotic proportional consensus, enabling multi-agent systems with various target tasks to achieve asymptotic proportional consensus, is presented. The theoretical results are validated through numerical examples.