中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2019, Vol. 38 ›› Issue (12): 1-7.DOI: 10.3969/j.issn.16740696.2019.12.01

• 交通+大数据人工智能 •    下一篇

复杂水域船舶避碰路径规划研究

谢新连,何平,何傲,辛剑英   

  1. (大连海事大学 综合运输研究所, 辽宁 大连 116026)
  • 收稿日期:2018-08-21 修回日期:2018-09-14 出版日期:2019-12-21 发布日期:2019-12-24
  • 作者简介:谢新连(1956—),男,辽宁大连人,教授,博士,主要从事交通运输规划与管理方面的研究。Email:xxlian@139.com。 通信作者:何平(1995—),男,四川巴中人,硕士研究生,主要从事交通运输规划与管理方面的研究。Email:455832384@qq.com。
  • 基金资助:
    中央高校基本科研业务费专项资金资助项目(3132016358);国家重点研发计划资助项目(2017YFC0805309)

Collision Avoidance Path Planning of Ships in Complicated Water Areas

XIE Xinlian, HE Ping, HE Ao, XIN Jianying   

  1. (Integrated Transport Institute, Dalian Maritime University, Dalian 116026, Liaoning, P. R. China)
  • Received:2018-08-21 Revised:2018-09-14 Online:2019-12-21 Published:2019-12-24

摘要: 针对目标船和碍航障碍区同时存在的复杂水域船舶避碰问题,在考虑《国际海上避碰规则》的前提下,结合避让目标船舶的预测危险区(PAD)理论、借鉴船舶领域相关成果进行的缓冲区分析以及切线图法对复杂水域进行环境建模,再用Dijkstra算法求得一条避开所有碍航区的初始路径;并利用船舶回转直径,基于改进的2turn方法对初始路径进行平滑处理,给出了避碰转弯角度和复航角度,其不仅能够平滑路径还能检验原路径的有效性;最后进行了仿真实验,验证了该方法的正确性与适用性;再与Maklink图结合Dijkstra和蚁群算法的路径规划方法进行对比分析,结果表明该方法在时间性能、路径整体的平滑性、路径优化程度等方面更具优势。

关键词: 船舶工程, 避碰路径规划, 复杂水域, 切线图法, Dijkstra算法, 2turn平滑

Abstract: For the problem of collision avoidance of ships in complicated water areas where both target ships and navigationobstructing areas exist at the same time, on the premise of considering the international regulations for preventing collisions at sea(COLREGS), the environment of complex waters was modeled by combining the predicted danger zone (PAD) theory of collision avoidance for targets ships, the buffer zone analysis referred to related achievements in the field of ship domain and the tangent graph method. And Dijkstra algorithm was used to find an initial path to avoid all the obstacle areas. Then, based on the improved 2turn method, the initial path was smoothed by using the ships turning diameter, and the collision avoidance turning angle and return angle were given, which could not only smooth the path but also test the effectiveness of the original path. Finally, the simulation experiment was carried out, and the correctness and applicability of the proposed method was verified. The path planning methods of Maklink graph, Dijkstra and ant colony algorithm were compared and analyzed. The results show that the proposed method has more advantages in time performance, smoothness of the whole path and the degree of path optimization.

Key words: ship engineering, collision avoidance path planning, complicated water areas, tangent graph method, Dijkstra algorithm, 2turn smoothing

中图分类号: