| 进化粒子滤波算法及其应用 |
| |
| 引用本文： | 莫以为,萧德云.进化粒子滤波算法及其应用[J].控制理论与应用,2005,22(2):269-272. |
| |
| 作者姓名： | 莫以为 萧德云 |
| |
| 作者单位： | 清华大学,自动化系,北京,100084 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划)资助项目(2002AA412510;2002AA412420). |
| |
| 摘 要： | 样本贫化现象是应用粒子滤波算法的一个主要障碍,对估计长时间维持不变量的影响尤为严重.通过分析产生该现象的原因,本文引入进化规划算子构成进化粒子滤波算法,增加样本集的多样性而缓解样本贫化现象的影响,改善其估计与跟踪能力,仿真结果显示所提出的算法是可行的.
|
| 关 键 词： | 粒子滤波算法 样本贫化 进化规划 状态估计 |
| 文章编号： | 1000-8152(2005)02-0269-04 |
| 收稿时间： | 9/3/2003 12:00:00 AM |
| 修稿时间： | 2004/5/31 0:00:00 |
Evolutionary particle filter and its application |
| |
| MO Yi-wei,XIAO De-yun.Evolutionary particle filter and its application[J].Control Theory & Applications,2005,22(2):269-272. |
| |
| Authors: | MO Yi-wei XIAO De-yun |
| |
| Affiliation: | Department of Automation,Tsinghua University,Beijing 100084,China |
| |
| Abstract: | Sample impoverishment phenomenon is a main handicap to particle filter application,especially in those cases to estimate the parameter that remains constant for a long time.Based on the analysis of the causes of sample impoverishment,the evolutionary particle filter is proposed,in which evolutionary programming is introduced.The improved approach relieves the effect caused by samples impoverishment through ameliorating the diversity of samples set.Simulation results demonstrate the feasibility of proposed evolutionary particle filter. |
| |
| Keywords: | particle filter samples impoverishment evolutionary programming state estimation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《控制理论与应用》浏览原始摘要信息 |
|
点击此处可从《控制理论与应用》下载全文 |
|
