随着航运业的飞速发展，海上交通日益繁忙，海上交通环境变得更为复杂，船舶碰撞事故的发生率也有所增加。因此，如何有效制定避碰策略，防止船舶碰撞事故的发生，充分保障船舶的航行安全，成为船舶自动避碰领域的研究重点。基于此，本文提出一种基于扩展式博弈的多船协商避碰算法，并构建一个综合考虑安全性、合规性和经济性3个因素的船舶避碰收益函数，以辅助船舶做出既安全又经济的避碰决策；同时，本文设计一种基于船舶安全会遇领域和速度障碍法（Velocity Obstacle, VO）的剪枝优化方法，旨在提高避碰算法的决策效率与实时性。为了验证所提算法的有效性，本文对典型的三船会遇场景进行仿真实验。仿真结果表明，本文所提算法能够有效应用于多船避碰问题。
With the rapid development of the shipping industry, the increasingly busy maritime traffic environment has become more complex, which has increased the incidence of ship collision accidents. Therefore, how to effectively formulate collision avoidance strategies to prevent ship collision accidents and fully ensure the navigation safety of ships has become the focus of research in the field of automatic collision avoidance of ships. Based on this, this paper proposes a multi-ship negotiation collision avoidance algorithm based on extensive game model, and constructs a collision avoidance payoff function of ships that comprehensively considers the three factors of safety, compliance of COLREGs and economy, so as to assist ships in making safe and economical avoidance decisions. At the same time, a pruning optimization method based on ship safety encounter domain and VO is designed to improve the decision-making efficiency and real-time performance of the collision avoidance algorithm. In order to verify the effectiveness of the algorithm, a simulation is carried out on the typical three-ship encounter scenario. The simulation results show that the proposed algorithm can be effectively applied to the problem of multi-ship collision avoidance.
2025,47(1): 76-82 收稿日期：2024-3-13
DOI：10.3404/j.issn.1672-7649.2025.01.014
分类号：U675
基金项目：国家自然科学基金资助项目（52371369）；福建省自然科学基金资助项目（2022J01323）；福建省科技计划对外合作项目（2023I0019）；福建省科技厅智能船舶关键技术试验与检测公共平台（3502ZCQXT2021007）
作者简介：庞樨(1996-),女,硕士研究生,研究方向为船舶避碰
参考文献：
[1] 中华人民共和国海事局. 事故调查报告[EB/OL]. https://www.msa.gov.cn/page/outter/shiguyufang.jsp, 2023-12-1/2024-3-1.
[2] LISOWSKI J. Multi-step matrix game with the risk of ship collision[J]. Risk Analysis IV: Simulation and Hazard Mitigation, 2004: 669-680.
[3] LISOWSKI J. The dynamic game models of safe navigation[J]. International Journal on Marine Navigation and Safety of Sea Transportation, 2007(1): 11-18.
[4] SZLAPCZYNSKI R. Evolutionary sets of safe ship trajectories: a new approach to collision avoidance[J]. Journal of Navigation, 2011, 64(1): 169-181.
[5] SZLAPCZYNSKI R. Evolutionary approach to ship's trajectory planning within traffic separation schemes[J]. Polish Maritime Research, 2012, 19(1): 11-20.
[6] SZTAPCZYNSKI R. Evolutionary sets of safe ship trajectories with speed reduction maneuvers within traffic separation schemes[J]. Journal of Navigation, 2013, 66(1): 65-81.
[7] SZLAPCZYNSKI R. Evolutionary planning of safe ship tracks in restricted visibility[J]. Journal of Navigation, 2015, 68(1): 39-51.
[8] 涂伊婧. 动态博弈框架下的多船避碰算法研究[D]. 武汉: 武汉理工大学, 2022.
[9] 侯建军, 许北平, 杨宝璋. 矩阵对策在船舶避碰中的应用研究[C]//船舶避让与安全操作. 大连: 大连海事大学航海学院, 2007.
[10] 王刚. 船舶扩展式博弈避碰决策系统建立与仿真研究[D]. 大连: 大连海事大学, 2014.
[11] 孔祥生, 卜仁祥, 刘勇. 基于扩展博弈理论的船舶自动避碰决策系统[J]. 计算机仿真, 2019, 36(5): 154-158+268.
[12] 欧阳旭东, 支云翔, 王腾飞, 等. 基于扩展式动态博弈的多船避碰决策模型[J]. 中国安全科学学报, 2020, 30(1): 128-135.
[13] 周江华, 吴春杰. 船舶碰撞危险度模型的构建[J]. 宁波大学学报(理工版), 2004(1): 61-65.
[14] DAVIS P V, DOVE M J, STOCKEL C T. A computer simulation of marine traffic using domains and arenas[J]. The Journal of Navigation, 1980, 33(2): 215-222.
[15] 郑中义, 吴兆麟. 船舶避碰决策[M]. 大连: 大连海事大学出版社, 2000.
