Optimal Control for fully-actuated Surface Vessel Systems

Abstract

This paper deals with the design of an optimal tracking control for fully-actuated Surface Vessel Systems with completely unknown dynamics. A feed-forward term in proposed controller is introduced for obtaining the corresponding autonomous tracking error model. An integral reinforcement learning (IRL) is then developed to solve Hamilton-Jacobi-Bellman (HJB) equation in optimal control term. The convergence of the proposed technique to the analytical solution of HJB equation is guaranteed. Additionally, the trajectory tracking effectiveness is also mentioned. Simulation studies are given to evaluated the quality of the proposed method.