Consensus of Singular Multi-agent Systems Based on Networked Predictive Control

Abstract

This brief studies the causal consensus problem of singular (descriptor) multi-agent systems with networked communication delays and agents described by general singular systems. For the studied systems, only the information of outputs are available through the network. An observer-based networked predictive control scheme is employed to compensate for the communication delays actively. Furthermore, based on the output feedback, observer and the networked predictive control scheme, a novel protocol is proposed in this paper. By using the tools of graph, algebra and singular system theory, the necessary and sufficient conditions are given for the existence of the proposed protocol to solve the considered consensus problem. The given conditions depend on not only the topologies of singular multi-agent systems but also the structure properties of each agent dynamics. Moreover, a consensus algorithm is proposed to design the novel observer-based predictive protocol. A numerical example demonstrates the effectiveness of compensation for networked delays using the provided consensus algorithm.