This article investigates the distributed fixed-time event-triggered consensus problem for linear multi-agent systems with input delay. The Artstein-Kwon-Pearson reduction approach is used to convert the delay-dependent system to a delay-free one. Based on the delay-free system, two distributed event-triggered protocols are designed based on local information exchange to address the leaderless consensus and the leader-follower consensus problems. In the proposed consensus protocols, both continuous controller updates and continuous communication as well as the Zeno behavior are successfully excluded. Detailed convergence analysis demonstrates that the closed-loop system is fixed-time stable regardless of the initial conditions.Finally, some simulation examples are provided to validate the theoretical results.
K E Y W O R D Sevent-triggered consensus, fixed-time convergence, input delay, multi-agent systems
INTRODUCTIONRecent years have witnessed considerable developments of consensus of multi-agent systems. 1 The driving force behind the research on this problem partly comes from its wide applications in sensor networks, 2 UAV formation, 3 and salvo attack of missiles. 4 The leaderless consensus and the leader-follower consensus are two fundamental problems in this area, which have been widely investigated in the existing works [5][6][7][8] and are also the focus of this article. Convergence time is an important performance index for multi-agent consensus from a practical point of view. Indeed, designing consensus protocols with finite-time stability, instead of asymptotic stability, contributes to better performance in higher precision, stronger robustness, and faster convergence rate. 9 However, lack of uniform boundedness of the settling time function regardless of the initial conditions restricts the practical applications of the existing finite-time consensus algorithms. 10,11 As an extension of the finite-time stability, the fixed-time stability, first referred in Reference 12, guarantees the settling time to be globally uniformly bounded regardless of the initial conditions. To this end, much effort has been made toward achieving fixed-time multi-agent consensus for first-or second-order systems. [13][14][15] Moreover, output-feedback based distributed algorithms were designed in References 16-18 to achieve the fixed-time consensus of second-order systems. Compared with first-or second-order systems, it is known that high-order systems are more attractive but technically challenging. With this in mind, the authors of References 19-22 considered the fixed-time consensus problem for multi-agent systems with high-order integrator-type dynamics, and the results concerning robust fixed-time consensus with parametric uncertainties were presented in References 23,24. In addition, the bipartite fixed-time consensus problem 2526