Applying the stereo-vision detection technique to the development of underwater inspection task with PSO-based dynamic routing algorithm for autonomous underwater vehicles

This research is an extension of image detection technique for obstacle-avoidance of autonomous underwater vehicles (AUVs) by applying the BK triangle sub-product of fuzzy relations. According to the concept of stereo-vision detection technique, the obstacles as well as offshore structures can be reconstructed by depth images. By obtaining the depth information in the space, the optimal route can be evaluated combining PSO (Particle Swarm Optimization)-based dynamic routing algorithm. In this study, the graphical language, LabVIEW (Laboratory Virtual Instrument Engineering Workbench), is used to simulate the AUV's inspection task in the offshore wind farm. The interface shows the pose, trajectories, perspectives and real-time series of 6-Degrees of Freedom (DOF) motion for the AUV. In the existence of obstacles, the AUV is found to conduct inspection tasks of the offshore wind farm with feasible routes by considering minimum time and energy consumption successfully. In summary, the stereo-vision detection technique with PSO-based dynamic routing algorithm is not only beneficial to optimize feasible routes but also identify features of objects for the purpose tracking and obstacle-avoidance more efficiently.