航空学报 > 2023, Vol. 44 Issue (11): 327561-327561   doi: 10.7527/S1000-6893.2022.27561

基于回访机制的无人机集群分布式协同区域搜索方法

文超1, 董文瀚2, 解武杰2, 蔡鸣2(), 刘日3   

  1. 1.空军工程大学 研究生院,西安 710038
    2.空军工程大学 航空工程学院,西安 710038
    3.空军哈尔滨飞行学院 理论训练系,哈尔滨 150000
  • 收稿日期:2022-06-01 修回日期:2022-06-30 接受日期:2022-09-29 出版日期:2023-06-15 发布日期:2022-10-14
  • 通讯作者: 蔡鸣 E-mail:caiming1124@sina.com
  • 基金资助:
    陕西省自然科学基础研究计划(2022JQ-584)

Distributed cooperative area search method for UAV swarms based on revisit mechanism

Chao WEN1, Wenhan DONG2, XIE Wujie2, Ming CAI2(), Ri LIU3   

  1. 1.Graduate School,Air Force Engineering University,Xi’an 710038,China
    2.College of Aeronautical Engineering,Air Force Engineering University,Xi’an 710038,China
    3.Department of Theory Training,Air Force Harbin Flight Academy,Harbin 150000,China
  • Received:2022-06-01 Revised:2022-06-30 Accepted:2022-09-29 Online:2023-06-15 Published:2022-10-14
  • Contact: Ming CAI E-mail:caiming1124@sina.com
  • Supported by:
    Basic Research Project in Natural Science of Shaanxi Province(2022JQ-584)

摘要:

为高效引导无人机(UAV)集群搜索未知任务区域内的动态目标,同时兼顾最大化覆盖搜索效能,提出一种回访机制驱动的UAV集群分布式协同搜索决策(RM-DCSD)算法。首先,基于栅格化方法构建了包含3种属性的综合态势信息图模型及其更新机理,为UAV进行实时在线搜索决策奠定基础;其次,以最大化搜索效率为优化目标,同时兼顾UAV的飞行安全与能耗代价,建立了UAV搜索效能函数,在此基础上,基于滚动优化思想进一步构建了UAV局部有限时域滚动优化模型;然后,综合考虑动目标的实际搜索需求以及传感器虚警和漏检情况,分别设计了信息素引导的回访机制与权系数动态切换引导的回访机制;接着,借鉴分布式模型预测控制思想,设计了基于信息融合的UAV集群分布式协同搜索决策机制,在确保集群分布式协同最优决策的基础上实现了对UAV成员态势信息图的解耦式更新,进一步增强了系统鲁棒性;最后,通过数值仿真实验对所提算法的有效性进行全面验证。仿真结果表明,RM-DCSD算法对动态未知搜索环境表现出良好的适应性,能够在引导UAV集群对未知区域进行最大化覆盖搜索的同时,通过回访机制驱动,有效兼顾对地面动目标的搜索需求。

关键词: 无人机集群, 协同区域搜索, 回访机制, 分布式模型预测控制, 信息融合

Abstract:

To enable Unmanned Aerial Vehicle (UAV) swarms to search for the dynamic targets efficiently in an unknown mission area while maximizing coverage of search, this paper proposes a Revisit Mechanism-driven Distributed Cooperative Search Decision (RM-DCSD) algorithm for UAV swarms. Firstly, a comprehensive situational information map model with three attributes and its updating mechanism are constructed based on the gridding method, laying the foundation for the UAV to make real-time online search decisions. Secondly, to maximize search efficiency and take into account the flight safety and energy cost of the UAV, a UAV search effectiveness function is established. On this basis, a UAV local finite-time domain rolling optimization model is constructed based on the thought of rolling optimization. Thirdly, considering the actual search demand for moving targets as well as the false alarm and missed detection of sensors, a pheromone-guided revisit mechanism and a weight coefficient dynamic switching-guided revisit mechanism are designed respectively. Then, drawing on the thought of distributed model predictive control, a distributed cooperative search decision mechanism for UAV swarms based on information fusion is designed, which achieves decoupled update of the situational information maps of UAV members on the basis of ensuring the distributed cooperative optimal decision of swarms, and further enhances the system robustness. Finally, the effectiveness of the proposed algorithm is verified by numerical simulation experiments. The results show that RM-DCSD performs good adaptability to dynamic unknown search environments, and can effectively consider the search requirements of ground moving targets through the revisit mechanisms while enabling UAV swarms to maximize coverage of search for unknown areas.

Key words: UAV swarm, cooperative area search, revisit mechanism, distributed model predictive control, information fusion

中图分类号: