Map based path-planning and guidance scheme of an AUV for inspection of artificial structures

This paper proposes a path-planning and guidance method of an autonomous underwater vehicle (AUV) to perform detailed inspection of underwater structures. The targets are support legs of on-water platforms consisting of cylindrical rods. Underwater inspection of such kind of structures is important for maintenance, security and environmental assessment. Autonomous underwater vehicles (AUVs) have a potential to make these tasks more inexpensive and reliable, compared with conventional methods. The path-planning method generates a set of waypoints for an AUV to follow the surface of the target with a constant distance, based on the pre-given information about the configurations of both the target and the vehicle. The guidance method actually controls the vehicle to follow the waypoints, assuming the position and attitude of the vehicle is precisely estimated in real-time. By combining these methods with the localization method we had proposed, an AUV can perform fully autonomous observation of underwater structures if only their configuration is known. The performance of the proposed method was verified though tank experiments using the AUV Tri-Dog 1. A simple model of a jacket was installed in a test tank and the vehicle succeeded in observing the target.