Observer based event-triggered leader-follower consensus of multi-agent systems

Input-observer event-based consensus in linear multi-agent systems

“…In this paper, an event triggering input observer is proposed, which does not need continuous measurements. Compared with the existing results in Zhang et al (2014), Yu et al (2018), Zhang et al (2016), Liu et al (2016), Wang et al (2017a), Cheng and Li (2019), Wang et al (2017b), and many other papers, although the control signal is updated intermittently, the observer needs continuous measurement to estimate states. In contrast, in our manuscript in the section ''Distributed event-triggered control with intermittent communication,'' there is no need for continuous communication.…”

“…To compare with the existing articles, an input observer is designed for the first time in linear MAS. In most existing papers (Cheng and Li, 2019;Liu et al, 2016;Lu and Li, 2016;Xing and Deng, 2018;Wang et al, 2017aWang et al, , 2017bYu et al, 2018;Zhang et al, 2016Zhang et al, , 2020, observers are designed to estimate states' values, and those values are sent to the neighbors. As explained in the introduction, input observer in MAS with the same observer gain will cause faster input estimation convergence, and consequently faster consensus is achieved.…”

“…Although similar problems have been studied before, event-based input observer is proposed for the first time. There are two main features in this manuscript: First, in most papers on EB MAS with linear dynamics like Yu et al (2018), Zhang et al (2016), Lu and Li (2016), Xing and Deng (2018), Liu et al (2016), Wang et al (2017a), Cheng and Li (2019), Wang et al (2017b), andZhang et al (2020), observers are designed to estimate states' values and those values are sent to the neighbors, while other observers used in EB MAS documents estimate either the leader's states, or disturbances. In this paper, motivated by Trentelman et al (2013), an input observer for MAS is designed where we do not need the estimate of the neighbors' states, the leader's state or the disturbance.…”

“…Since the controller is fed by the observer, the consensus is reached faster if the observer converges faster. Moreover, in most related existing articles like Yu et al (2018), Zhang et al (2016), Liu et al (2016), and Cheng and Li (2019), although the control signal is updated in triggering times, observers need continuous communication to estimate the states; in contrast, in this article, both continuous and intermittent communication are considered. The main contribution of this paper is to show that MAS's states will converge to a common value, by only using neighbors' output.…”

“…Next, leader–follower MASs are a vital part of networked systems, where control objectives can be achieved by controlling a leader. In Zhang et al (2016), an event-triggered consensus of a first order leader–follower MAS with undirected communication graph is investigated. In this paper, followers estimate leader’s state to track its output.…”