Cooperative Neural Network Control for Uncertain Non-affine Multi-agent Systems via an Adaptive Distributed Observer

This paper considers cooperative tracking problem of the multi-agent system modeled by a class of uncertain non-affine nonlinear systems and an autonomous nonlinear leader system. In this paper, an adaptive distributed observer was firstly designed, which could estimate the leader's signal to various followers. Then, the non-affine system is transformed into an affine form by a low-pass filter. Finally, a simulation example is provided to verify the present controller. This controller provides three interests: it calculates the observer gain on-line to process communication barriers; the non-affine multi-agent system is transformed into affine form to overcome the control difficulty of non-affine multi-agent system; the proposed control scheme only needs two online approximators to estimate the unavailable states.