Neural-Network-Based Event-Triggered Secure Synchronization Control for a Class of Uncertain Nonlinear MASs Under FDI Attacks

In this paper, an observer-based neural network control scheme is proposed for the secure synchronization control problem of multi-agent systems (MASs) with unmodeled nonlinear dynamics under FDI attacks. A more general network attack scenario in which both control signals and measurement outputs can be tampered with by malicious attackers is considered during information transmission. A closed-loop observer is designed using a separation method for the case of an unmeasurable state of MASs. To save network resources and reduce the impact of FDI attacks, an event-triggered mechanism is introduced to reduce the transmission of the measurement output signals. In addition the designed event-triggered mechanism excludes Zeno behavior.