Route planning for autonomous vessels based on improved artificial fish swarm algorithm

• Published in: Ships and offshore structures Vol. 18; no. 6; pp. 897 - 906
• Main Authors: Zhao, Liang, Wang, Fang, Bai, Yong
• Format: Journal Article
• Language: English
• Published: Cambridge Taylor & Francis 03.06.2023; Taylor & Francis Ltd
• Subjects: AFSA; Algorithms; ASV; Comparative analysis; Comparative studies; Ecosystem disturbance; GNC system; GNC systems; Path planning; Probability theory; Route planning; Sea trials; Surface craft
• ISSN: 1744-5302; 1754-212X
• DOI: 10.1080/17445302.2022.2081423

Path planning is one of the key technologies in the research of autonomous surface vessels (ASVs). In this paper, an improved artificial fish swarm algorithm (IAFSA) is proposed. The algorithm is modified from four perspectives: (1) A directional operator is introduced to improve the efficiency. (2) To avoid local optimum, a probability weight factor is proposed to adjust the frequency of executing random behaviour. (3) An adaptive operator has been applied aims at better convergence performance. (4) The waypoint modifying path smoother is used to improve the path quality. A comparative study has been carried out between IAFSA and the other state-of-the-art algorithms, and the results indicate that the proposed algorithm outperforms the others in both efficiency and path quality. Finally, IAFSA is integrated into the GNC system in a model ship. A computer-based sea trial around the Nan Hai area has been conducted, and environmental disturbances including wind, waves, and currents are considered. The results have shown that IAFSA is suitable for practical application.