Distributed adaptive average-consensus control for a class of second-order nonlinear multi-agent system using neural network

Abstract

In this paper, an effective distributed adaptive control method is proposed to solve the second-order average-consensus problem for multi-agent system with nonlinear dynamics. First of all, the distributed adaptive law is put forward in order to solve the problem that each agent can only use its neighboring agents information. Then, the basic idea of backstepping is used to solve the second-order dynamics. Since the neural network has the favorable approximation capability, the uncertain nonlinear dynamics in this paper is solved by it. By combining the aforementioned several methods, the designed controller for each agent can guarantee the average-consensus behavior could be obtained and all the signals here are bounded. Finally, the effectiveness and feasibility of the proposed approach is illustrated by the simulation examples.