Obstacle avoidance method appropriate for the steep terrain of the deep seafloor

Autonomous underwater vehicles(AUVs) are vital tools for detailed ocean survey in recent years. Amongst them, hovering-type AUVs are specialized in optical observation using cameras and travel near the seafloor avoiding obstacles. Therefore, it is important to maintain proper distance between AUV and objects which should be approached to take pictures.The aim of our work is to enhance the robustness of the observation by AUVs by reducing false detection of obstacles and planning proper distance when avoiding obstacles. By hoveringtype AUV HOBALIN equipped with cameras and line lasers, obstacles are detected by simple image processing techniques and measured three-dimensional positions by light-sectioning method. A series of movement of the proposed method was confirmed through experiments in test water tank. Then the method could be implemented to AUVs and utilized at the actual sea surveys.