Asynchronous Event-Triggered Passive Consensus of Semi-Markov Jump Multiagent Systems With Two-Time-Scale Feature Under DoS Attacks

This article deals with the asynchronous event-triggered passive consensus issue for two-time-scale semi-Markov jump multiagent systems in continuous-time domain against denial-of-service (DoS) attacks. The multiagent systems are modeled with two-time-scale structures and the parameter changes are assumed to follow semi-Markov jump processes. For multiagent systems with two-time-scale structures, singular perturbed approaches are introduced to separate the coupled slow and fast dynamics of each agent. Then, double-rate sampling mechanisms are adopted for slow and fast dynamics with different evolution rates, and a novel asynchronous event-triggered scheme (AETS) with two independent triggering conditions is proposed. The asynchronous event-triggered scheme based on double-rate sampled-data reduces the transmission of unnecessary data and saves limited bandwidth resources by triggering independently on two time-scales. In addition, a consensus strategy is proposed to assure the multiagent system of achieving consensus with a specified passive performance level. Finally, the simulation example verifies that the asynchronous event-triggered scheme saves about 50% of network resources than the synchronous scheme under the premise of the same control effect.