Large-scale path planning for Underwater Gliders in ocean currents

Underwater gliders are a class of AUVs designed for high endurance over long distances, but their reduced velocity makes them more susceptible to ocean currents during deployment. Thus, feasible paths need to be generated through the ocean current field. This paper proposes a method for determining energy-optimal paths that account for the influence of ocean currents. The proposed technique is based on Rapidly-Exploring Random Trees (RRTs). Using real ocean current and bathymetry data, results produce comparable paths to grid based methods, and offer an improvement in terms of avoiding high-energy shallow regions. Future work will focus on heuristically biasing the RRT growth to further improve the generated paths, and implementation of the algorithms on a glider platform.