北京航空航天大学学报 ›› 2020, Vol. 46 ›› Issue (10): 1826-1833.doi: 10.13700/j.bh.1001-5965.2019.0405

• 论文 • 上一篇    下一篇

基于AIGWO-IMMUKF的目标跟踪算法

游航航1,2, 韩其松1, 余敏建1, 龙宏志3, 杨海燕1, 李朋永2   

  1. 1. 空军工程大学 空管领航学院, 西安 710051;
    2. 中国人民解放军 94498部队, 南阳 473000;
    3. 中国人民解放军 95380部队, 湛江 524000
  • 收稿日期:2019-07-19 发布日期:2020-10-29
  • 通讯作者: 韩其松 E-mail:afeu_yh@126.com
  • 作者简介:游航航 男,硕士,助理工程师。主要研究方向:航空兵指挥;韩其松 男,硕士,讲师。主要研究方向:航空兵指挥自动化;余敏建 男,硕士,教授。主要研究方向:航空兵指挥自动化;杨海燕 女,博士,副教授。主要研究方向:空天态势与威胁评估;李朋永 男,本科,工程师。主要研究方向:空域管理。
  • 基金资助:
    国家自然科学基金(61472441);装备预研领域基金(61403110304);空军工程大学校长基金(XZJY2018031)

Target tracking algorithm based on AIGWO-IMMUKF

YOU Hanghang1,2, HAN Qisong1, YU Minjian1, LONG Hongzhi3, YANG Haiyan1, LI Pengyong2   

  1. 1. Air Traffic Control and Navigation College, Air Force Engineering University, Xi'an 710051, China;
    2. Unit 94498 of the PLA, Nanyang 473000, China;
    3. Unit 95380 of the PLA, Zhanjiang 524000, China
  • Received:2019-07-19 Published:2020-10-29
  • Supported by:
    National Natural Science Foundation of China (61472441); Fund for Equipment Pre-research Field of China(61403110304); President's Fund of Air Force Engineering University (XZJY2018031)

摘要: 针对目标跟踪算法中滤波器选择和模型设计问题,提出了一种具有自适应性的交互式多模型无迹卡尔曼滤波(IMMUKF)目标跟踪算法。首先,介绍了IMMUKF的算法步骤;其次,提出运用改进的灰狼优化(IGWO)算法优化其中的滤波参数,通过构造调节因子建立了时变的Markov状态转移概率,形成了AIGWO-IMMUKF算法,并给出其算法流程;最后,将所提AIGWO-IMMUKF算法与传统算法在相同条件下进行仿真,得出位置、速度均方根误差曲线,以及时效性对比。结果表明,所提AIGWO-IMMUKF算法克服了传统IMMUKF算法的不足,提升了算法性能,精度和时效性都更优。

关键词: 目标跟踪, 无迹卡尔曼滤波(UKF), 交互式多模型(IMM), 灰狼优化(GWO), 滤波参数, 转移概率

Abstract: Aimed at the problem of filter selection and model design in target tracking algorithm, an adaptive Interacting Multiple Model-Unscented Kalman Filter (IMMUKF) target tracking algorithm is proposed. First, the algorithm steps of IMMUKF are introduced. Second, the Improved Grey Wolf Optimizer (IGWO) is proposed to optimize the filter parameters, and the time-varying Markov state transition probability is established by constructing the adjustment factor. Then, the AIGWO-IMMUKF algorithm is formed and its algorithm flowchart is given. Finally, the AIGWO-IMMUKF algorithm proposed in this paper and the traditional method are simulated under the same conditions, and the root mean square error curves of position and velocity as well as the timeliness comparison chart are obtained. The results show that AIGWO-IMMUKF algorithm overcomes the shortcomings of traditional IMMUKF, improves the performance of the algorithm, and has better accuracy and timeliness.

Key words: target tracking, Unscented Kalman Filter(UKF), Interacting Multiple Model(IMM), Grey Wolf Optimizer(GWO), filter parameter, transition probability

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发