Human-in-the-Loop Leader-Following Consensus Control for Nonlinear MASs Subject to Deception Attacks via Dynamic Self-Triggered Mechanism

This article investigates the human-in-the-loop self-triggered consensus control problem for nonlinear multiagent systems under unknown deception attacks and output constraints. Deception attacks are shown to degrade sensor accuracy, complicating the design of adaptive laws and control strategies. To mitigate these effects, a novel coordinate transformation is introduced. The local control and adaptive laws are adjusted to effectively counteract the negative influence of output constraints. In addition, Nussbaum-type functions are employed to compensate for unknown, time-varying control gain discrepancies caused by deception attacks. A dynamic self-triggering mechanism is further developed to reduce communication overhead imposed by bandwidth limitations. Based on the Lyapunov stability theory, it is rigorously proven that the proposed secure controller guarantees leader–follower consensus. The effectiveness of the control strategy is demonstrated through two simulations.