Efficient Multi-Task 3D Path Planning for AUV in Complex Marine Environment

This paper investigates the multi-task path planning problem for Autonomous Underwater Vehicle in complex three-dimensional marine environments. Traditional marine monitoring methods are costly and inefficient. A UV, with their high autonomy and flexibility, are crucial for seafloor surveys. To ensure safe navigation and task completion, this study combines Ant Colony Optimization, Particle Swarm Optimization, and Rapidly-exploring Random Tree algorithms for efficient path planning. Simulation experiments in a three-dimensional ocean environment with obstacles evaluated the performance of the hybrid algorithm, and its efficiency in path planning was verified through comparisons with PSO+RRT, GWO+RRT, and RRT. Results show that the hybrid algorithm effectively generates collision-free paths, enhancing global search capability, local optimization, and path feasibility while considering environmental factors like ocean currents, thereby improving A UV operational efficiency and range. This study offers a robust and efficient path planning solution, supporting future developments in autonomous navigation technologies and improving AUV efficiency and reliability in real-world applications.