An Efficient Algebraic Model Predictive Control (AMPC) for Unmanned Surface Vessels Path Following

The path following of an underactuated unmanned surface vessel (USV) in a complex marine environment requires minimizing both the track error and yaw angular velocity, whereas the underactuated characteristics of USV and variability of model parameters need to be considered. To achieve this, the problem of path tracking is modelled using a first-order Nomoto model in the Serret-Frenet coordinate system. An efficient algebraic model predictive control (AMPC) approach is employed to determine the optimal control law. This can significantly reduce the number of prediction points in traditional predictive control, simplify the computation of online optimization, and contribute to the implementation of robust control for USV path following. The results of simulation demonstrate that the developed controller achieves asymptotic stability with respect to the track error deviation, heading angle deviation, and heading angle speed. The performance of the controller in terms of good stability and robustness is also verified.