A potential field approach for reactive navigation of autonomous sailboats

Navigation techniques for autonomous sailboats are faced with two inherent difficulties. The uncontrollable and partially unpredictable nature of thrust forces on one hand and the complex kinematics of sailboats on the other hand. This paper proposes a new reactive navigation approach, based on artificial potential fields, that addresses these two problems simultaneously. Environment and specific sailboat navigation constraints are represented by a local potential built around the vehicle location. Changes of wind direction and detected obstacles affect this periodically updated potential, which guarantees the real-time computation of a feasible heading for the boat. Numerical simulations are presented and validate the proposed algorithm under various wind conditions. Field trials eventually illustrate the efficiency of this navigation technique using a reduced-scale autonomous sailboat prototype.

► We propose a novel reactive navigation technique for autonomous sailboats.
► Our navigation approach is based on artificial potential fields.
► The specific sailboat kinematics is integrated into a local potential field built around the robot location.
► Our method has been tested in simulation and has been validated on a lake using a reduced-scale autonomous sailboat.