Consensus control for a class of second-order multi-agent systems: An iterative learning approach

This paper studies the distributed consensus control problem for a class of leader-follower multi-agent systems with double integrator dynamics by using an iterative learning approach. Consensus errors of individual agents are considered as the anticipation in time and a distributed iterative learning protocol for an undirected networks with a fixed topology is proposed to guarantee that the follower agents track the leader perfectly in finite time. The dynamic of the leader agent is assumed to be time-varying and its state information is available to only a portion of the followers. A sufficient condition is obtained for the consensus control problem of the multiagent system, and a numerical example is provided to show the effectiveness of the proposed method.