作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (3): 149-151. doi: 10.3969/j.issn.1000-3428.2011.03.053

• 人工智能及识别技术 • 上一篇    下一篇

RBPF粒子滤波在目标跟踪中的应用研究

赵 丰1,2,汤 磊3,陈国友1,赵宗贵4   

  1. (1. 解放军理工大学指挥自动化学院,南京 210007;2. 海南省军区,海口 570236; 3. 南京陆军指挥学院,南京 210045;4. 中国电子科技集团公司第28研究所,南京 210014)
  • 出版日期:2011-02-05 发布日期:2011-01-28
  • 作者简介:赵 丰(1972-),男,工程师、博士研究生,主研方向:信息融合,指挥自动化;汤 磊,讲师、博士;陈国友,副教授; 赵宗贵,研究员、博士生导师
  • 基金资助:
    国家部委基金资助项目

Research on Application of RBPF in Target Tracking

ZHAO Feng 1,2, TANG Lei 3, CHEN Guo-you 1, ZHAO Zong-gui 4   

  1. (1. Institute of Command and Automation, PLA University of Science & Technology, Nanjing 210007, China; 2. Hainan Province Military District, Haikou 570236, China; 3. Nanjing Army Command College, Nanjing 210045, China; 4. The 28th Research Institute, China Electronics Technology Group Corporation, Nanjing 210014, China)
  • Online:2011-02-05 Published:2011-01-28

摘要: 针对雷达目标跟踪中的某一类非线性问题,传统处理方法都是先行线性化再进行处理,但当线性化后的测量噪声相关性较大时则无法满足要求。为此,应用RBPF粒子滤波进行研究。在仿真实验中对RBPF多权值的情况进行探讨,提出一种可行的处理方法。对RBPF在噪声相关性较大时的性能进行分析,并讨论其在时间推移时的相对估计误差变化情况。

关键词: RBPF粒子滤波, Kalman滤波, 目标跟踪

Abstract: Rao-Blackwellised Particle Filtering(RBPF) is a class of particle filter. A particular nonlinear radar target tracking problem is analyzed in detail, this kind of problem is often dealt with linearization and it is failed when correlation between measurement noise is too big. In this simulation test, multi-weight of RBPF is discussed and a doable way is proposed, the performance of RBPF under much correlation of noise is analyzed, relative estimate error of RBPF is also discussed when time processes.

Key words: Rao-Blackwellised Particle Filtering(RBPF), Kalman filtering, target tracking

中图分类号: