Nonlinear path tracking control of underactuated surface vessel by LSSVM-based identification

A fully instrumented surface vessel involving underactuated structure was designed. The problem of straight line course tracking of the underactuated surface vessel at a constant forward speed is addressed. The Serret-Frenet frame is used to define the tracking error, therefore the position tracking errors could be stabilized by stabilizing the single distance error. Least Squares Support Vector Machines (LSSVM) is proposed to identify surface vessel based on free-running model tests and full-scale trials, respectively. A tracking controller based on State-Dependent-Riccati Equation (SDRE) optimizing control method is proposed to solve the challenging problem. The characteristic of SDRE guarantees the stability of the closed-loop system. Simulation and experiments results are presented to validate the effectiveness of the proposed method.