Signed-average consensus for networks of agents: a nonlinear fixed-time convergence protocol

Abstract

This paper addresses the consensus problem of multi-agent systems where the interactions between agents are not only cooperative but also antagonistic. A distributed nonlinear protocol is proposed to guarantee the agents to reach agreement bipartitely regarding a signed-average quantity that depends on both the initial states of agents and the topology structures of multi-agent systems. Moreover, this signed-average consensus can be achieved within a fixed settling time which is not dependent upon the initial states of agents but can be estimated by the design parameters of the presented protocol. Numerical simulations are performed to illustrate the fixed-time convergence performance of the established signed-average consensus results.