Output-based dynamic event-triggered consensus control for linear multiagent systems

Abstract

We apply dynamic event-triggered control to a general linear multiagent system. On the basis of output feedback, a decentralized dynamic event-triggered controller is proposed for each agent to achieve consensus. No agent requires continuous communication, and it is sufficient for an agent to continuously observe its own states and broadcast them to its neighbors at the event time. The event frequency can be greatly reduced without appreciable loss of consensusability compared with the case of conventional static event-triggered controllers. Each event occurs with appropriate timing because we use dynamically varying thresholds for event triggering. The effectiveness of the proposed method is confirmed by simulation. Furthermore, the consensusability and event frequency are compared between static and dynamic event-triggered controllers in numerical simulation.