Adaptive distributed consensus tracking control for uncertain nonlinear multi-agent systems in pure-feedback form
Xiaocheng Shi, Shengyuan Xu · 2016
This paper investigates an adaptive distributed consensus tracking problem of uncertain nonlinear multi-agent systems in pure-feedback form on a directed graph. The unknown continuous nonlinear functions induced from the controller design procedure are approximated by Radial basis function neural networks (RBFNNs). Based on the distributed dynamic surface control technique, the problem of “explosion of complexity” in the traditional backstepping design is eliminated by introducing a first-order filtering. Differing from the existing results in the literature, the proposed new design approach only needs one learning parameter to be updated on line for M pure-feedback nonlinear followers. It is also shown that the proposed consensus controllers can guarantee cooperatively semi-global uniform ultimate boundedness (CSUUB) of all the signals, and the consensus errors converge to an adjustable neighborhood of the origin.