An obstacle avoidance system for autonomous underwater vehicles: A reflexive vector field approach utilizing obstacle localization

Now that Autonomous Underwater Vehicles (AUVs) are exploring more challenging terrain than ever, the need for an obstacle avoidance system has become apparent. Obstacle avoidance systems (OAS) in unmanned systems are far from new. However, adapting existing methodologies to AUVs presents a new set of challenges. Underwater obstacle sensing relies on limited and often noise prone acoustic range sensors, vehicles have limited maneuverability, and minimizing interference from other acoustic devices is challenging. Last year, International Submarine Engineering Ltd (ISE) tackled the task of adding an OAS to its line of Explorer AUVs. Using a combination of vector field, reactive, and object mapping algorithms, a robust and configurable obstacle avoidance system was developed.