Distributed Event-Triggered Consensus of General Linear Multiagent Systems Under Directed Graphs

This article investigates the consensus problem of general linear multiagent systems under directed communication graphs with event-triggered mechanisms. First, a novel distributed static event-triggered mechanism with a state-dependent threshold is proposed to solve the consensus problem, both with a positive lower bound on the average time interval of the communication among agents and updates of controllers. Thus, the Zeno behavior is excluded for communication among agents and controller updates. Next, to further reduce the frequencies of communication among agents and updates of controllers, a distributed dynamic event-triggered mechanism is introduced. By applying the static and dynamic mechanisms, the problem can be addressed with the reduced use of system resources compared with that in most existing control algorithms. Finally, numerical simulations are presented to verify the effectiveness of the results.

I. Introduction

The cooperative control of multiagent systems has attracted considerable attention from researchers in various fields due to its widespread applications, such as in flocking, cooperative unmanned air vehicles, formation control, and distributed sensor networks [1]–[4]. In the cooperative control of multiagent systems, the consensus problem is a critical issue, which means that all agents reach an agreement with respect to certain quantities of interest through interaction among agents. For the consensus problem of multiagent systems, other interesting works also exist, such as consensus problems in the presence of DOS attacks [5] and finite/fixed-time consensus problems [6]–[8].