| 引用本文: | 陈辉,刘雅婷,张双庆,韩崇昭.多扩展目标跟踪中基于加权最优子模式分配距离的传感器管理方法[J].控制理论与应用,2022,39(5):887~896.[点击复制] |
| CHEN Hui,LIU Ya-ting,ZHANG Shuang-qing,HAN Chong-zhao.Sensor management method based on weighted optimal sub-pattern assignment distance in multiple extended target tracking[J].Control Theory and Technology,2022,39(5):887~896.[点击复制] |
|
| 多扩展目标跟踪中基于加权最优子模式分配距离的传感器管理方法 |
| Sensor management method based on weighted optimal sub-pattern assignment distance in multiple extended target tracking |
| 摘要点击 1532 全文点击 478 投稿时间：2021-02-01 修订日期：2022-01-12 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 10.7641/CTA.2021.10112 |
| 2022,39(5):887-896 |
| 中文关键词 传感器管理 随机有限集 加权最优子模式分配 多扩展目标跟踪 离差 |
| 英文关键词 sensor management random finite set W–OSPA multiple extended target tracking dispersion |
| 基金项目 国家自然科学基金项目(62163023, 61873116, 61763029), 国防基础科研项目(JCKY2018427C002), 甘肃省教育厅产业支撑计划项目(2021CYZC– 02)资助. |
|
| 中文摘要 |
| 针对多扩展目标跟踪优化中的传感器管理问题, 本文基于随机有限集(RFS)提出一种基于加权最优子模式
分配(W–OSPA)距离的传感器管理方法. 首先, 在部分可观测马尔可夫决策过程(POMDP)理论框架下, 给出基于逆
威沙特椭圆分布假设的多扩展目标跟踪传感器管理的基本方法. 其次, 利用W–OSPA距离求解多扩展目标后验分布
的离差, 以离差最小化获得最优化的多扩展目标状态估计设计传感器管理策略. 最后, 通过推导给出的离差数值求
解方法进行最优传感器管理方案的决策. 仿真实验基于OSPA距离评价验证了所提传感器管理方法相对于其他算法
对多扩展目标的质心运动状态和椭圆形状都有更好的估计效果. |
| 英文摘要 |
| To solve the problem of sensor management in multiple extended target tracking optimization, on the basis
of the random finite set (RFS), a sensor management algorithm is proposed through the weighted optimal sub-pattern
assignment (W–OSPA) distance. First, in the framework of partially observable Markov decision process (POMDP) theory,
the basic sensor management method is given in multiple extended target tracking based on the ellipse assumption of inverse
Wishart distribution. Then, the W–OSPA distance is used to solve the dispersion of the multiple extended target posterior
distribution and correspondingly the sensor management strategy is designed to obtain the optimal multiple extended target
state estimation by minimizing dispersion. After that, the decision of optimal sensor management scheme is made through
the proposed numerical solution method of dispersion. The simulation experiments based on the OSPA distance evaluation
verify that the proposed sensor management algorithm has a better estimation effect on the centroid motion state and the
ellipse shape of multiple extended target than other algorithms. |
|
|
|
|
|