Nonlinear adaptive path following control of an autonomous surface vehicle: Theory and experiments

Abstract

Autonomous surface vehicles (ASVs) have been widely used for military, civilian and academic applications because of their low developing costs and high executing efficiency. This paper investigates the path following control problem for a low-cost ASV without velocity measurement and feedback control. A Serret-Frenet coordinate frame, which originates at a reference point on the reference path, is utilized to describe path following errors and the corresponding dynamic model. The reference point moves along the path with a designed rate, which produces an additional control input to solve the singularity problem presented in the classic method. The moving velocity of the reference point and the desired yaw angle are appropriately designed to ensure the path following errors to be globally asymptotically stable with the estimated velocity and side-slip angle by using an adaptation mechanism. The steering control of ASVs is designed using an adaptive sliding mode control technique by dealing with the uncertain parameters in yaw dynamics. The simulations and experiments with linear and circular reference paths demonstrate the feasibility and effectiveness of the proposed path following controller.