Nonlinear Model Predictive Control for Tracking of Underactuated Vessels under Input Constraints

In this paper, a nonlinear model predictive control (NMPC) is presented for position and velocity tracking of underactuated surface vessel with input constraints. A hree-degree-of-freedom (3-DOF) dynamic model is used with only two control variables: namely, surge force and yaw oment. Without frame transformation, a nonlinear, but nvex, opti-mization problem is formulated to minimize the deviation of the vessel states from a time varying reference generated over a fi-nite horizon by a virtual vessel with the same dynamics. A real-time C-code is generated, using ACADO toolkit and qpOASES solver, with multiple shooting technique for discretization andGauss-Newton iteration algorithm, which is computationally efficient, thus enabling real-time implementation of proposed technique. MATLAB simulations is used to assess the validity of the proposed technique.