系统工程与电子技术 ›› 2019, Vol. 41 ›› Issue (10): 2243-2251.doi: 10.3969/j.issn.1001-506X.2019.10.13

• 系统工程 • 上一篇    下一篇

空天高速目标探测跟踪传感器资源调度模型与算法

高嘉乐, 邢清华, 梁志兵   

  1. 空军工程大学防空反导学院, 陕西 西安 710051
  • 出版日期:2019-09-25 发布日期:2019-09-24

Multiple sensor resource scheduling model and algorithm for high speed target tracking in aerospace

GAO Jiale, XING Qinghua, LIANG Zhibing   

  1. Air and Missile Defense College, Air Force Engineering University, Xi’an 710051, China
  • Online:2019-09-25 Published:2019-09-24

摘要: 针对空天高速目标跟踪中,传感器资源调度任务无时间等待、传感器资源匹配形式多样、观测时间碎片化等问题,提出了多源异构传感器调度多目标优化模型和求解该模型的多目标柔性果蝇算法。针对传感器调度时间碎片化问题,采用柔性分割调度时间,设计了目标-时间-传感器三维编码方式;为了避免相似个体交叉进化造成搜索陷入局部最优,提出基于个体特征的交叉操作和变异操作;针对进化过程中个体产生碎片时间、重复观测等问题,提出柔性调整操作。最后通过两个案例测试,对提出的模型和算法进行仿真验证,实验证明算法能够合理地求解多源异构传感器资源调度问题,在收敛性和分布性方面优于对比算法。

关键词: 果蝇算法, 多目标, 传感器资源调度, 高速目标

Abstract: For the high-speed target tracking in aerospace, sensor resource scheduling has problems such as lack of task latency time, multiple matching of sensor resources and observation time fragmentation. To solve these problems, this paper proposes a multi-source heterogeneous sensor scheduling multi-objective optimization model and a multi-objective flexible fruit fly algorithm for solving the model. In order to solve the problem of time fragmentation, the target-time-sensor three-dimensional coding method is designed by using flexible segmentation scheduling time. To avoid the search to fall into local optimum caused by evolution of similar individuals, the crossover operator and mutation operator based on individual features are proposed. Focusing on the problems of time fragmentation and repeated observation during evolution, a flexible adjustment operator is presented. Finally, the model and algorithm proposed in this paper are verified by two test problems. The experimental results show that the proposed algorithm can solve the multi-source heterogeneous sensor resource scheduling problems reasonably, and the proposed algorithm has better properties than the comparison algorithm in convergence and distribution.


Key words: fruit fly algorithm, multi-objective, sensor resource schedule, high speed target