Lagrangian coherent structure assisted path planning for transoceanic autonomous underwater vehicle missions

Abstract

Transoceanic Gliders are Autonomous Underwater Vehicles (AUVs) for which there is a developing and expanding range of applications in open-seas research, technology and underwater clean transport. Mature glider autonomy, operating depth (0–1000 meters) and low energy consumption without a CO2 footprint enable evolutionary access across ocean basins. Pursuant to the frst successful transatlantic glider crossing in December 2009, the Challenger Mission has opened the door to long-term, longdistance routine transoceanic AUV missions. These vehicles, which glide through the water column between 0 and 1000 meters depth, are highly sensitive to the ocean current feld. Consequently, it is essential to exploit the complex space-time structure of the ocean current feld in order to plan a path that optimizes scientifc payof and navigation efciency. This letter demonstrates the capability of dynamical system theory for achieving this goal by realizing the real-time navigation strategy for the transoceanic AUV named Silbo, which is a Slocum deep-glider (0–1000m), that crossed the North Atlantic from April 2016 to March 2017. Path planning in real time based on this approach has facilitated an impressive speed up of the AUV to unprecedented velocities resulting in major battery savings on the mission, ofering the potential for routine transoceanic long duration missions.