Distributed adaptive tracking control for a class of uncertain multiagent systems

Abstract

In this paper, a new distributed adaptive control for a class of multiagent systems with model uncertainties is proposed to track the desired trajectory specified by an informed agent. It is assumed that there also exist parametric uncertainties and unknown dynamics with the informed agent. Linearly parameterized approximators are employed to estimate these unknowns and only part of agents needs to have the access to the state value of the informed agent in the proposed design. A baseline design is first provided for first-order multiagent models, and then extension is made to second-order multiagent systems using adaptive backstepping techniques. In essence, the proposed distributed adaptive controller can be taken as a dynamic one in the sense that the control itself relies on agent's own state values while the states of neighbors are used in the adaptive laws for parameter estimation. The asymptotical consensus tracking is rigorously proved under the assumption of the connected sensing/communication topology among agents. Simulation results are included to illustrate the proposed new design.