海上自主航行船舶的避碰决策应做到安全、合理、有效，避碰决策应能在满足海上真实航行场景需求的同时，还应尽可能多地响应多种不同的船舶会遇局面，这对避碰决策算法的完备性提出很高要求。考虑多航行特征，构建初始避碰决策空间，设置多船会遇约束、避碰规则约束和偏航约束，并引入有限状态机模型，提出海上自主航行船舶避碰策略。通过仿真发现，在两船会遇中OS将其航线向右舷侧调整30°，有效避免进入目标船舶TS的领域；在多船会遇中，OS根据规则要求采取让路行动，与TS保持在安全的会遇距离之外。考虑多航行特征的海上自主航行船舶避碰策略综合考量TS的各种潜在响应行为，能够实现两船会遇和多船会遇的有效避障，更贴近海上实际避碰场景的需求，具有一定的实践应用价值。
The collision avoidance decision-making of autonomous ships at sea should be safe, reasonable and effective, and the collision avoidance decision-making should be able to meet the demands of the real navigation scenarios at sea while responding to as many different ship encounters as possible, which puts forward high requirements for the completeness of the collision avoidance decision-making algorithm. The initial collision avoidance decision space is constructed by considering multiple sailing characteristics, setting the constraints of multi-ship encounter, collision avoidance rules and yaw constraints, and introducing the finite state machine model to propose collision avoidance strategies for autonomous ships at sea. Through simulation, it is found that the OS adjusts its course to the starboard side by 30° in the two-ship encounter, effectively avoiding entering the domain of the target ship TS. In the multi-ship encounter, the OS takes give way action according to the rule requirements, keeping out of the safe encounter distance with TS. The collision avoidance strategy for autonomous ships at sea considering multiple navigational features integrates various potential response behaviours of TS, and is able to achieve effective obstacle avoidance in two-vessel encounters and multi-vessel encounters, which is closer to the needs of actual collision avoidance scenarios at sea, and has certain practical application value.
2025,47(7): 6-12 收稿日期：2024-5-27
DOI：10.3404/j.issn.1672-7649.2025.07.002
分类号：U675
基金项目：国家重点研发计划课题资助项目（2022YFC3006001）；自主创新基金资助项目（104972024KFYd0029）
作者简介：罗曼尔(2002-),女,本科,武汉理工大学,研究方向为水上交通安全、海上防污染
参考文献：
[1] LEE J H, PARK J D. Collision evasive action timing for MASS using CNN-LSTM-based ship trajectory prediction in restricted area[J]. Ocean Engineering, 2024, 294(1): 116766.
[2] 周怡, 袁传平, 谢海成, 等. 基于DDPG算法的游船航行避碰路径规划[J]. 中国舰船研究, 2021, 16(6): 19-26.
[3] 孙硕, 杨少龙, 向先波, 等. 基于动态领域势场法的船舶避碰路径规划[J]. 水下无人系统学报, 2023, 31(5): 679-686.
[4] 刘钊, 周壮壮, 张明阳, 等. 基于双延迟深度确定性策略梯度的船舶自主避碰方法[J]. 交通信息与安全, 2022, 40(3): 60-74.
[5] 张金奋, 王伦巍, 吴达, 等. 基于自适应步长快速搜索随机树算法的船舶避碰路径规划[J]. 中国航海, 2023, 46(4): 147-153.
[6] 王晋鑫, 张政. 基于人工势场法和模型预测控制的船舶变速避碰研究[J]. 科技资讯, 2023, 21(21): 46-49.
[7] WANG T, YOUN H I. Development of a graph-based collision risk situation model for validation of autonomous ships’ collision avoidance systems[J]. Journal of Marine Science and Engineering, 2023, 11(11): 224-228.
[8] KIM I. Systematization of legal procedures for collision avoidance between a fully autonomous ship and a traditional manned ship[J]. Journal of Marine Science and Engineering, 2023, 11(10): 145-149.
[9] ZHENG K, ZHANG X, WANG C, et al. Adaptive collision avoidance decisions in autonomous ship encounter scenarios through rule-guided vision supervised learning[J]. Ocean Engineering, 2024, 297(1): 117096.
[10] 赵兴亚, 贺益雄, 黄立文, 等. 复杂环境下船舶慢速域自主航行决策方法[J]. 华中科技大学学报(自然科学版), 2024, 52(3): 113-120.
[11] 徐卫名. 基于海事规则的中型无人艇避碰路径规划算法分析[J]. 珠江水运, 2023(24): 94-96.
[12] 张可, 黄立文, 贺益雄, 等. 基于航迹推演的船舶动态智能避碰方法[J]. 中国航海, 2023, 46(4): 20-29.
[13] 崔浩, 张新宇, 王警, 等. 自主船舶与有人驾驶船舶动态博弈避碰决策[J]. 中国舰船研究, 2024, 19(1): 238-247.
[14] 关巍, 崔哲闻, 罗文哲. 基于改进PPO算法的船舶自主避碰决策[J]. 大连海事大学学报, 2023, 49(4): 28-36.
[15] 赵贵祥, 王晨旭, 王贺平, 等. 改进速度障碍法的无人艇局部路径规划[J]. 系统工程与电子技术, 2023, 45(12): 3975-3983.
[16] 马杰, 苏钰栋, 熊勇, 等. 基于速度障碍和人工势场的受限水域船舶避碰决策方法[J]. 中国安全科学学报, 2020, 30(11): 60-66.
[17] 黄立文, 李浩宇, 梁宇, 等. 基于操纵过程推演的船舶可变速自动避碰决策方法[J]. 交通信息与安全, 2021, 39(6): 1-10.
[18] 汪栋, 张杰, 金久才, 等. 一种基于有限状态机模型的局部转向避碰路径规划算法[J]. 海洋科学, 2018, 42(1): 119-127.
[19] 李啸雨, 许劲松, 杨荣武. 多船会遇自主避碰算法[J]. 船舶工程, 2018, 40(9): 67-71+101.
[20] 张英俊, 翟鹏宇. 海运船舶自主避碰技术研究进展与趋势[J]. 大连海事大学学报, 2022, 48(3): 1-11.
