3D Detection and Tracking of Mooring Lines of Floating Offshore Wind Turbines by Autonomous Underwater Vehicle

This paper introduces an innovative method for inspecting mooring lines of Floating Offshore Wind Turbines (FOWTs) using an Autonomous Underwater Vehicle (AUV) equipped with a tilt-controlled Multibeam Imaging Sonar (MBS). The approach aims to enable the AUV to accurately estimate the position of the mooring lines and safely track them. This method overcomes the limitations in traditional inspection techniques and lays the groundwork for the integration of additional methods for more detailed inspections. With the tilt angle of the MBS and a pre-trained YOLO model, the AUV estimates 3D positions of the mooring chains with sonar imagery. The tracking method is designed to maintain safe distance to the chains by dynamically adjusting surge and sway velocities, based on the real-time detections. Through continuous tracking, the AUV is capable of reconstructing the whole structure of the mooring line with its position data estimated by dead-reckoning. The proposed method was tested using a hovering type AUV, Tri-TON, in a sea trial at an FOWT, Hibiki, in Kitakyushu, Japan. The experiment validated the AUV's effectiveness in tracking the mooring lines for 343 seconds and successfully reconstructing their entire structure. Moreover, the experimental results indicate a need for further tasks, particularly in enhancing the AUV's positional accuracy and in conducting a thorough analysis of the chains' physical characteristics to facilitate a more comprehensive evaluation of their condition.