中国舰船研究2025，Vol.20Issue(1)：172-180,9.DOI:10.19693/j.issn.1673-3185.03929
基于改进DDPG算法的无人船自主避碰决策方法
Autonomous decision-making method of unmanned ship based on improved DDPG algorithm
摘要
Abstract
[Objectives]To enhance the safety and efficiency of maritime traffic,this paper proposes an autonomous collision avoidance decision-making method for unmanned ships based on an enhanced Deep De-terministic Policy Gradient(DDPG)algorithm.[Methods]In order to address the issues of low data utiliza-tion and poor convergence in traditional DDPG algorithms,we employ Priority Experience Replay(PER)to dynamically adjust experience priority,reduce sample correlation,and utilize the Long Short-Term Memory(LSTM)network to improve the algorithm convergence.Based on the domain knowledge of ships and adher-ing to the International Regulations for Preventing Collisions at Sea(COLREGs),a model for determining meeting situations and a novel set of reward functions that consider urgent scenarios when other ships fail to comply with the COLREGs are introduced.Generalization experiments are conducted involving two-ship and multi-ship encounters to validate the effectiveness of the proposed method.[Results]As the experimental results demonstrate,compared to traditional DDPG algorithms,our improved approach enhances the conver-gence speed by approximately 28.8%.[Conclusions]The trained model enables autonomous decision-mak-ing and navigation while ensuring compliance with the COLREGs,thereby providing valuable insights for in-telligent decision-making in the field of maritime transportation.关键词
无人船/深度确定性策略梯度算法/自主避碰决策/优先经验回放/国际海上避碰规则/避碰Key words
unmanned vehicles/deep deterministic policy gradient(DDPG)algorithm/autonomous colli-sion avoidance decision-making/priorit-ized experience replay(PER)/international regulations forprevent-ing collisions at sea(COLREGs)/collision avoidance分类
交通工程引用本文复制引用
关巍,郝淑慧,崔哲闻,王淼淼..基于改进DDPG算法的无人船自主避碰决策方法[J].中国舰船研究,2025,20(1):172-180,9.基金项目
国家自然科学基金资助项目(51409033,52171342) （51409033,52171342）
