Event-Triggered Resilient Consensus Control for Multiagent Systems Under Asynchronous Optimal DoS Attacks

Abstract

This article investigates the design of asynchronous optimal denial-of-service (DoS) attacks targeting consensus mechanisms in multiagent systems (MASs). Current research predominantly relies on random DoS attack strategies that fail to account for adversarial intentionality and high-intensity behavioral patterns inherent to cyber threats. To address this limitation, we propose an asynchronous optimal DoS (AODoS) attacks algorithm specifically designed for energy-constrained adversaries. The algorithm strategically optimizes channel selection to maximize consensus error. To counteract such attacks, secure control strategies integrating event-triggered and self-triggered mechanisms is developed to achieve resilient bipartite consensus of MASs. Theoretical analysis derives sufficient conditions for ensuring consensus convergence under AODoS attacks. Numerical simulations validate the effectiveness of the proposed strategies in mitigating high-intensity DoS attacks.