航空学报 > 2020, Vol. 41 Issue (S2): 724294-724294   doi: 10.7527/S1000-6893.2020.24294

基于群智机理的集群防碰撞控制

姜龙亭1,2, 魏瑞轩2, 张启瑞3, 王栋2   

  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. 中国人民解放军95561部队, 日喀则 857000
  • 收稿日期:2020-05-26 修回日期:2020-06-01 发布日期:2020-06-24
  • 通讯作者: 姜龙亭 E-mail:kgdjltsmile@163.com
  • 基金资助:
    科技部重点项目"新一代人工智能"(2018AAA0102403);国家自然科学基金(61573373)

Anti-collision control of UAVs based on swarm intelligence mechanism

JIANG Longting1,2, WEI Ruixuan2, ZHANG Qirui3, WANG Dong2   

  1. 1. Graduate College, Air Force Engineering University, Xi'an 710038, China;
    2. Aeronautics Engineering College, Air Force Engineering University, Xi'an 710038, China;
    3. Troop 95561 of PLA, Rikaze 857000, China
  • Received:2020-05-26 Revised:2020-06-01 Published:2020-06-24
  • Supported by:
    Key Projects of the Ministry of Science and Technology of "New Generation Artifical Intelligence"(2018AAA0102403);National Natural Science Foundation of China (61573373)

摘要: 在执行作战任务以及民用中,无人机集群由于具备高可靠性和高效费比的特点而受到广泛关注;随着城市低空空域的开放,作战环境的动态复杂性剧增,对无人机集群防碰撞提出了严峻的挑战。针对无人机集群执行任务时的编队个体之间防碰撞以及遭遇突发障碍时的集群防碰撞问题,提出了基于群智机理的集群防碰撞控制方法。通过引入力学概念,设计了集群内部吸引力、排斥力以及编队构型力的模型;通过集群个体间的信息共享以及类脑知识发育,构建了基于群信息共享和类脑反射的无人机防碰撞模型。仿真结果表明:所提集群控制方法提高了集群的集结效率,而且在遭遇突发障碍时能够规避动态障碍,再次完成编队重构;群智机理的引入缩短了无人机集群面临突发障碍时的反应时间。

关键词: 群智机理, 无人机集群, 防碰撞, 类脑反射, 无人机

Abstract: UAV clusters with high reliability and an efficient cost ratio have attracted extensive attention in both combat missions and civil use. The opening of urban low-altitude airspace and the dramatically increasing dynamic complexity of combat environments impose severe challenges for UAV cluster collision prevention. To solve the problem of collision prevention in both formation and encountering with sudden obstacles, a cluster collision prevention control method based on the swarm intelligence mechanism is proposed. The models of attraction, repulsive force and formation configuration force within the cluster are designed by introducing the concept of mechanics, and a collision prevention model of UAVs based on group information sharing and brain-like knowledge development is constructed through information sharing among individuals in the cluster and the development of brain-like knowledge. The simulation results show that the proposed cluster control method can improve the clustering efficiency, avoid suddenly encountered dynamic obstacles, and complete formation reconstruction. The introduction of swarm intelligence mechanism can shorten the response time of UAV clusters in face of sudden obstacles.

Key words: swarm intelligence mechanism, UAV cluster, anti-collision, brain-like reflection, UAVs

中图分类号: