A Path Planning Method for Unmanned Surface Vessels in Dynamic Environment

• Published in: International journal of control, automation, and systems Vol. 22; no. 4; pp. 1324 - 1336
• Main Authors: Yu, Jiabin, Chen, Zhihao, Zhao, Zhiyao, Xu, Jiping, Lu, Yang
• Format: Journal Article
• Language: English
• Published: Bucheon / Seoul Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers 01.04.2024; Springer Nature B.V; 제어·로봇·시스템학회
• Subjects: Algorithms; Control; Engineering; Mechatronics; Nodes; Obstacle avoidance; Ocean currents; Path planning; Regular Papers; Robotics; Sea vessels; 제어계측공학; DWA algorithm; COLREGs; path planning; RRT algorithm; space-varying ocean current
• ISSN: 1598-6446; 2005-4092
• DOI: 10.1007/s12555-022-1172-7

A path planning method for unmanned surface vessels (USV) in dynamic environment is proposed to address the impact of dynamic environments on path planning results and the lack of dynamic obstacle avoidance capabilities. First, the considering ocean current rapidly exploring random tree (RRT*) (COC-RRT*) algorithm was proposed for global path planning. The RRT* algorithm has been enhanced with the integration of the virtual field sampling algorithm and ocean current constraint algorithm. The COC-RRT* algorithm optimizes the global planning path by adjusting the path between the parent nodes and child nodes. Second, according to the limitations of the International Regulations for Preventing Collisions at Sea (COLREGs), the improved dynamic window approach (DWA) is applied for local path planning. To enhance the ability of avoid dynamic obstacles, the dist function in the DWA algorithm has been improved. Simulation experiments were conducted in three scenarios to validate the proposed algorithm. The experimental results demonstrate that, in comparison with other algorithms, the proposed algorithm effectively avoids dynamic obstacles and mitigates the influence of the space-varying ocean current environment on the path-planning outcome. Additionally, the proposed algorithm exhibits high efficiency and robustness. The results verified the effectiveness of the proposed algorithm in dynamic environments.