Leader-following consensus of multi-agent systems with time delays by fully distributed protocols

This paper considers the consensus problem of linear multi-agent systems with large input and communication delays. Provided that only the relative output among agents is available and the delays are constant and exactly known, a novel reduced-order observer-based finite dimensional protocol is proposed to solve the considered problem. If it is further assumed that the systems only contain input delays, then the delays are allowed to be time-varying and unknown. It should be noted that the delays under consideration are allowed to be arbitrarily large yet bounded and the proposed protocols do not need the a priori global information of the communication topology and thus are robust to changes in communication topology. Finally, simulation results on spacecraft formation flying system verify the effectiveness of the proposed approach.