J4 ›› 2014, Vol. 41 ›› Issue (6): 31-36+75.doi: 10.3969/j.issn.1001-2400.2014.06.006

• 研究论文 • 上一篇    下一篇

一种新的进化裂变粒子滤波算法

王泽玉;李明;张鹏   

  1. (西安电子科技大学 雷达信号处理国家重点实验室,陕西 西安  710071)
  • 收稿日期:2013-10-11 出版日期:2014-12-20 发布日期:2015-01-19
  • 通讯作者: 王泽玉
  • 作者简介:王泽玉(1990-),女,西安电子科技大学硕士研究生,E-mail: beidou13579@163.com.
  • 基金资助:

    国家自然科学基金资助项目(61271297,61272281, 61301284);博士学科点科研专项基金资助项目(20110203110001);国家部委预研基金资助项目(9140A0702****DZ01001)

Novel particle filter algorithm based on evolution fission

WANG Zeyu;LI Ming;ZHANG Peng   

  1.  (National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2013-10-11 Online:2014-12-20 Published:2015-01-19
  • Contact: WANG Zeyu

摘要:

针对经典自举粒子滤波中的重要性函数选取和重采样所导致的样本枯竭问题,提出了一种基于进化裂变的改进粒子滤波算法.该算法首先采用无迹卡尔曼滤波算法产生重要性函数,然后对重要性采样粒子进行裂变通过进化策略更新粒子集以增加粒子多样性,从而克服经典自举滤波重采样过程中的粒子退化问题.仿真实验表明,该算法能有效地提高跟踪精度, 跟踪性能优于经典粒子滤波算法.

关键词: 粒子滤波, 重采样, 重要性函数, 进化裂变, 无迹卡尔曼滤波算法

Abstract:

In order to overcome the choice of the importance function and the sample impoverishment after resampling, an improved particle filter algorithm based on evolution fission is presented. The algorithm uses the UKF to generate the importance function and updates the samples based on evolution fission, which could increase the diversity of the samples and overcome the degeneration of the typical particle filter. Simulation results prove that the presented algorithm has a high tracking accuracy and a good tracking performance compared to the typical particle filter.

Key words: particle filter, resampling, the importance proposal distribution, evolution fission, unscented particle filter(UKF)

中图分类号: 

  • TN953
Baidu
map