Mean-square Consensus of Heterogeneous Continuous-time Multi-agent Systems with Measurement Packet Losses*

In this paper, the mean-square output consensus problem for continuous-time heterogeneous multi-agent systems (MASs) with different measurement packet losses under directed communication topologies is studied. Previous work mainly focus on the design of optimal estimators for discrete-time MASs, but there are few systematic studies on continuous-time systems with measurement packet losses. In order to solve this problem, an optimal estimator for continuous-time MASs is developed through discretization, and the asymptotic stability of the optimal estimator is discussed through theoretical derivation, so as to effectively deal with the uncertainty caused by measurement packet losses. Based on the designed optimal estimator, a distributed dynamic feedback control protocol is proposed, and sufficient conditions for continuous-time systems to achieve mean-square output consensus are obtained, which depends on system dynamics, communication topology and packet loss probability. Finally, in order to verify the validity of the proposed method and theoretical results, a simulation example is given.