A two level approach for underwater path planning

This paper proposes a path planning strategy for an underwater vehicle moving in a 3D environment. A free path from an initial point to a goal that takes into account the a priori knowledge on the environment (obstacles and current profile) and copes, in real time, with unexpected obstacles is derived. The planner has a two level structure that combines a high level (HLP) with a low level planner (LLP). The HLP uses the a priori knowledge on the environment to derive a free geometric path defined by the initial location, a set of intermediate way-points and the goal location. This path optimizes the vehicle energy profit from the currents. The LLP implements a potential field based technique to generate the set of vehicle configurations between two consecutive way-points