Probability-guaranteed secure consensus control for time-varying stochastic multi-agent systems under mixed attacks

In this paper, the secure consensus control issue is investigated for a class of discrete time-varying stochastic multi-agent systems (MASs) subject to cyber-attacks. In order to give a comprehensive characterization of malicious threats against communication networks, a generic model is presented to take into account both random false data injection attacks (FDIAs) and replay attacks. The main objective of the problem under study is to design a control protocol via output feedback such that, despite the existence of mixed attacks, all the individual agents can be driven to reside within a desired ellipsoidal region in a pre-specified probability. Sufficient conditions are provided for the existence of the requested controller and the feedback gains are formulated in terms of the solution to certain matrix inequalities. Within the established framework, two optimization problems are considered with the aim to ensure the sub-optimal consensus performances from different perspectives. Finally, a simulation example is employed to illustrate the validity of the proposed control scheme. (C) 2022 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.

Automated summary

In this paper, the secure consensus control issue is investigated for a class of discrete time-varying stochastic multi-agent systems subject to cyber-attacks. A generic model is presented to account for random false data injection attacks and replay attacks. The objective is to design a control protocol that can drive all individual agents to reside within a desired ellipsoidal region despite the existence of mixed attacks. The paper provides sufficient conditions for the existence of the requested controller and formulates the feedback gains in terms of certain matrix inequalities.