Nonlinear URANS model for path-following control problem towards autonomous marine navigation under wave conditions

An accurate prediction of the performance of maritime autonomous surface ships (MASS) to follow a predefined path in waves is of critical importance for ensuring safe autonomous marine navigation. Traditional methods for the study of path-following problems are based on simplified mathematical ship models and are incapable of precisely resolving the complicated interactions between the hull, propeller, rudder, and incident waves under path-following control. In the present study, a CFD-based dynamic model for path-following problems is developed for autonomous marine navigation by means of a fully nonlinear unsteady Reynolds-Averaged NavierStokes (RANS) solver combined with the Line-of-Sight (LOS) guidance law. Fortunately, an unsteady RANS solver is capable of incorporating viscous and turbulent effects and the free surface resolution critical to path-following problems, allowing a better prediction of a ship's path-following performance in waves. To obtain practical insight into the performance of the ship executing path-following control for autonomous navigation in waves, a comprehensive analysis of path-following tasks in waves covering the whole range of important wave directions and wave heights was carried out in this study. The numerical results demonstrated that when the ship performed the path-following manoeuvre under the beam and quartering seas, the greatest oscillatory deviation from the predefined path was observed due to the asymmetric pressure distribution on the ship hull caused by the incident waves. It was also confirmed that wave heights strongly affected the trajectories experienced by the ship performing the path-following task in the beam and oblique seas; the deviation from the predefined path was found to increase with the increase in the wave height. As high-performance computational resources become increasingly available, the proposed CFD method will provide an accurate and efficient way to predict the performance of autonomous surface ships performing path-following tasks in waves, providing a valuable contribution to enhancing the safety of autonomous marine navigation.