Autonomous navigation of ships by combining optimal trajectory planning with informed graph search

Autonomous trajectory generation plays an essential role in the navigation of vehicles space as well terrestrial scenarios, i.e. air, on solid ground, or water. For latter, ships ports has specific challenges since ship dynamics are highly nonlinear with limited agility, while manoeuvre is limited. Nevertheless, for providing support to humanly designed control strategies, autonomously generated trajectories have not only be feasible, collision-free but shall also optimal respect time and effort. This article presents a novel approach autonomous planning basis precomputed connectable segments, so-called motion primitives, A*-search algorithm. Sequences primitives provide initial guess subsequent optimization by which found even terrains many obstacles. We illustrate different scenarios.