Abstract
Based on the basic principles of genetic algorithm and movement features of Autonomous Underwater Vehicles (AUVs), this paper proposes a system that optimizes the path to a target point and velocity along it. Multi-objective optimization by the evaluation functions of location and speed precisions at the stopping point, and consumed energy is conducted using ε-limitation and two-time selection of pareto-optimal. The proposed system can include collision avoidance performance. The efficiency of the system is demonstrated in the simulation based on the AUV 'Twin-Burger'.
