航空学报 > 2023, Vol. 44 Issue (S2): 729758-729758   doi: 10.7527/S1000-6893.2023.29758

基于速度障碍的多无人船协同避碰

何信1,2, 石宗英1(), 钟宜生1   

  1. 1.清华大学 自动化系,北京 100084
    2.中国运载火箭技术研究院 研发中心,北京 100076
  • 收稿日期:2023-10-24 修回日期:2023-11-21 接受日期:2023-12-11 出版日期:2023-12-20 发布日期:2023-12-18
  • 通讯作者: 石宗英 E-mail:szy@mail.tsinghua.edu.cn
  • 基金资助:
    国家重点研发计划(2020YFF0400101)

Multi⁃USV cooperative collision avoidance based on velocity obstacle

Xin HE1,2, Zongying SHI1(), Yisheng ZHONG1   

  1. 1.Department of Automation,Tsinghua University,Beijing 100084,China
    2.Research and Development Center,China Academy of Launch Vehicle Technology,Beijing 100076,China
  • Received:2023-10-24 Revised:2023-11-21 Accepted:2023-12-11 Online:2023-12-20 Published:2023-12-18
  • Contact: Zongying SHI E-mail:szy@mail.tsinghua.edu.cn
  • Supported by:
    National Key Research and Development Program of China(2020YFF0400101)

摘要:

无人船在航行时会遇到无法预知的障碍物和其他船只,并面临与其碰撞的风险。在繁忙的交通海域和复杂的环境中,这样的风险就更严峻。基于《国际海上避碰规则》对于船只避碰操纵行为的规定并进行多船避碰情况下的规则拓展,提出了一种多无人船的协同避碰方法。所提方法结合了速度障碍法以及各会遇情况下船只避碰行为的规定,综合考虑了动态船只与其他障碍的存在,创新地将实际的避碰问题转化为凸优化问题,为无人船提供实时的避碰参考信号,相较目前研究处理多船协同避碰问题效果更好、效率更高。相较已有研究进一步进行了复杂避碰情景下的较长时间仿真验证,以验证所提方法的可靠性。本文方法在复杂多船避碰情形下能够快速给出协同避碰指令,保障无人船安全航行,拥有较好的效果。

关键词: 无人船, 协同避碰, 速度障碍法, 凸优化, 碰撞

Abstract:

USVs will encounter unpredictable obstacles and other vessels and face the risk of collision when navigating. Such risks are exacerbated in heavy water traffic. Based on COLREGs, a cooperative collision avoidance method for multi-USVs is designed in this paper. This method combines the velocity obstacle method with the COLREGs, taking the existence of dynamic vessels and other obstacles into account. This method transforms the collision avoidance problem into a convex optimization problem, and provides collision avoidance reference signals in real-time. Simulation validations under various collision avoidance scenarios are carried out to show the reliability of the proposed method.

Key words: Unmanned Surface Vehicle (USV), cooperative collision avoidance, velocity obstacle method, convex optimization, collision

中图分类号: