Secure Fault-Tolerant Control of Leaderless Multiagent Systems via a Dynamic Event-Triggered Mechanism Under DoS Attacks
Abstract
This paper investigates the secure consensus problem for leaderless multiagent systems (MASs) under the joint influence of denial of service (DoS) attacks and intermittent actuator faults. Since DoS attacks and actuator faults can both influence the MASs' performance, it is necessary to explore the coexistence of these two phenomena. To solve this problem, a secure fault-tolerant control strategy is proposed. Firstly, an adaptive neural network controller based on backstepping is designed to deal with unknown nonlinearity and actuator faults. Secondly, a dynamic event-triggered (DET) mechanism is introduced to improve communication efficiency and robustness. The proposed DET strategy incorporates an attack detection logic and dynamically adjusts triggering thresholds according to system conditions. Additionally, a local error surface method is employed to achieve leaderless consensus without relying on reference signals. Under DoS attacks and actuator faults, the uniform ultimate boundedness of all closed-loop signals is rigorously established via Lyapunov analysis, and Zeno behavior is excluded. Simulation results demonstrate that the proposed approach ensures effective consensus while balancing communication economy and resilience against attacks and faults.