Observer-based adaptive consensus of intelligent vehicle convoys with directed and undirected networks subject to input saturation

Abstract

In this research work, the consensus control design of intelligent vehicle convoys with directed and undirected networks and third-order dynamic equation is studied. The engine saturation as an input saturation is involved in the dynamical model which will make the longitudinal dynamics of each following intelligent vehicle to be nonlinear. All followers track an actual leading intelligent vehicle that has no input. For some reasons such as reducing the communication burden, external attacks and sensor failure, it is assumed that some of the states of followers and the leader are not available. Three common network structures are considered in consensus design and analysis: some predecessors following, bi-directional and uni-directional. To estimate the inaccessible states of followers and the leader, two different observers combined with an output feedback controller are designed so that each follower can estimate its own and the leader's unknown states. Afterward, a linear output feedback law is introduced to attain the global asymptotic consensus. The Lyapunov theory is applied to prove that the output control protocol leads the proposed intended convoy to asymptotic stability in the linear and nonlinear input saturation regions. For all three structures, several examples are provided to evaluate the method of this article.