Semi-global bipartite consensus tracking of singular multi-agent systems with input saturation.

This paper investigates the problem of bipartite consensus tracking for a class of linear singular multi-agent systems under a signed graph topology, where the control input of each agent is subject to saturation. By exploiting a parametric algebraic Riccati equation (ARE)-based low-gain feedback approach, a static state feedback control protocol and a dynamic observer-based output feedback control protocol are developed utilizing local information. It is shown that under the presented protocols, semi-global bipartite consensus tracking can be achieved if the underlying signed communication graph is structurally balanced and contains a spanning tree rooted at the leader. A numerical example is finally given to illustrate the effectiveness of the theoretical results.