Practical consensus of networked agents with uncertain high-order-nonlinear dynamics

In this paper, the practical consensus problem of networked nonlinear heterogeneous agent with uncertain parameter as well as modeling error is considered under fixed sensing/communication topology. A graphic-theoretical approach combined with adaptive parameter estimation and robust controller is proposed such that the network can reach practical consensus under the graph topology condition mentioned in the paper. Compared with most existing work, the ultimate consensus state is independent of uncertainty and the heterogeneity of the agent. In spite of the practical consensus which means there is some difference between the ultimate state of agent, but the error can be minimized as small as possible. what´s more, the information needing to be transferred is very limited. This new result enables an adaptive robust controller(ARC)for consensus protocol for network of agent with high-order nonlinear dynamics and having modeling error. The designing procedure of the protocol is based on back-stepping, and its effectiveness is proven by Lyapunov Function. Finally examples are presented to illustrate the effectiveness of the proposed controller.