Energy-efficient path planning for fully propelled AUVs in congested coastal waters

Abstract

To navigate safely in littoral waters with complex currents and busy shipping activities, it is crucial to have small autonomous vehicle with full propulsion capability. Such systems typically have limited endurance. We therefore investigate the feasibility of improving their endurance using current-aware path planning algorithms that allow active propulsion to move the vehicle into favorable currents or to avoid obstacles. The strategy adopted is to minimize the use of propulsion while leveraging favorable currents as much as possible. We perform simulations using environmental data and operational constraints of autonomous vehicles. The current field is assumed to be time-invariant over the period of the mission. Simulation results for 40 randomly generated source-destination pairs in Singapore Strait are presented. The performance is quantified by comparing the energy consumption of the path generated against the shortest distance path. Simulation results show that we are able to save 30-90% energy if the vehicles are allowed to drift along with the current. When the minimum speed of the vehicle is constrained to 2.5 knots, the energy savings could range from a few percent to more than 50%, depending on the currents along the route. The expected energy saving is the largest when the vehicle is allowed to operate at speeds comparable to water current while the savings diminish when the vehicle is required to operate at higher speed. © 2013 IEEE.