Adaptive output consensus tracking of uncertain multi-agent systems

Abstract

In this paper, we consider the adaptive output consensus tracking of a class of nonlinear systems in their higherorder strict-feedback form with mismatched parametric uncertainties. The networked group of nonlinear systems has a time-varying reference as a virtual leader, whose output is available to partial agents in the network. The proposed approach features in a design procedure including distributed estimation of group reference and backstepping-based robust consensus tracking control. In the proposed consensus tracking design, each agent estimates the consensus reference through local communications with its neighbours, and makes distributed control decision based on the estimated consensus reference. The proposed approach allows robust consensus tracking in the presence of mismatched model uncertainties, and the tracking performance is characterised by an index showing the effect of the unavailable reference for part of the networked agents. Sufficient conditions are given to ensure bounded output consensus tracking. Simulation results show satisfactory performances.