Leader-following event-triggered consensus control of MASs under false data injection attacks

This paper studies the event-triggered consensus problem of leader-following linear multi-agent systems (MASs) under stochastic false data injection attacks (FDIAs), where the false data is injected into the communication channels between the network agents. The randomly occurring stealthy FDIAs on the communication channels are modeled by the Bernoulli process distributions. By designing a distributed event-triggered consensus control strategy, the consensus problem of the MAS with one-to-all FDIA scenario is studied firstly, then it is extended to the more flexible one-to-one FDIA scenario with a maximum number of destroyed neighboring channels for each agent. The event-triggered mechanism of the state-dependent threshold is proposed and the auxiliary variable is introduced to further prevent Zeno behavior. By utilizing stochastic analysis method, two bounded consensus theorems for directed MAS networks are presented for the above two cases of attack signals respectively to guarantee the followers synchronize to the leader within a limited margin of error ultimately. Finally, simulation results are provided to illustrate the validity and feasibility of the proposed event-triggered consensus control strategies.