An obstacle avoidance algorithm for an autonomous underwater vehicle based on fuzzy-virtual potential function and forward looking sonar

Abstract

Obstacles avoidance problem plays a key role in the performance of the Autonomous Underwater Vehicles (AUVs). Obstacle detection is the first step in the process of crossing an obstacle. Sonar is the most used obstacles detection equipment in aquatic environments. In this research work, a forward looking sonar model is presented to calculate the distance and direction relative to the obstacle which is used in potential function to generate a virtual repulsion force to obstacles avoidance. The main path is due to the line of sight guidance method and appropriate side angle is generated by combining repulsive force in x-y plane to deviate from the main path and leads away from the obstacle. To optimize the side angle, the multiplication gain of distance and bearing weighting functions is applied based on fuzzy logic. The validity of the proposed obstacles avoidance algorithm is investigated in 6-DOF dynamical model of an AUV. The results indicate the appropriate performance of the proposed algorithm.

Keywords

• autonomous underwater vehicle
• artificial potential functions
• forward looking sonar
• fuzzy weighting function.