The optimal route planning for inspection task of autonomous underwater vehicle composed of MOPSO-based dynamic routing algorithm in currents

For supporting the dynamic routing plan more efficiently, this study has been established by integrating PSO (Particle Swarm Optimization) −based dynamic routing algorithm, self-tuning fuzzy controller, a stereo-vision detection technique and 6-DOF mathematical model into the inspection system of AUV (Autonomous Underwater Vehicle). Specifically, the PSO-based dynamic routing algorithm is modified by adopting the concept of Multi-Objective Particle Swarm Optimization (MOPSO), which is able to handle different weights of objectives in parallel. Therefore, a modular structure is applied to program design of the system by using the graphical language, LabVIEW^®, which is composed of 6-DOF motion module, a self-tuning fuzzy control module, a stereo-vision detection module, and a dynamic routing module. Performances resulted from the MOPSO-based dynamic routing algorithm would be discussed by conducting a series of inspection tasks in the imitated offshore wind farm. Additionally, selections of fixed weight and dynamic weight of MOPSO-based dynamic routing algorithm would be compared via Pareto frontiers for feasible solutions of both sailing time and energy consumption. Eventually, it is verified that the MOPSO-based dynamic routing algorithm in our system is not only able to estimate the feasible routes intelligently, but also identify features of underwater structures for the purpose of positioning.