| 观测值含有粗差的抗差粒子滤波算法 |
| |
| 引用本文： | 范澎湃,隋立芬,曹轶之.观测值含有粗差的抗差粒子滤波算法[J].测绘工程,2009,18(2):24-27. |
| |
| 作者姓名： | 范澎湃 隋立芬 曹轶之 |
| |
| 作者单位： | 信息工程大学,测绘学院,河南,郑州,450052 |
| |
| 摘 要： | 作为非线性滤波的代表，粒子滤波得到广泛应用。该算法通过随机生成的具有权重的样本（粒子）来计算后验概率密度。其中赋权的过程需要综合系统信息和观测信息，当观测值含有粗差时，会使粒子错误赋权影响滤波结果。提出一种新的基于抗差估计的抗差粒子滤波算法，通过计算粒子等价权，抑制观测值粗差的影响。模拟计算分析表明，当观测值含有粗差时，与标准粒子滤波相比，该方法能有效提高滤波精度。
|
| 关 键 词： | 粒子滤波 抗差估计 等价权 抗差粒子滤波 |
Dealing with the measurement outlier by robust particle filter |
| |
| FAN Peng-pai,SUI Li-fen,CAO Yi-zhi.Dealing with the measurement outlier by robust particle filter[J].Engineering of Surveying and Mapping,2009,18(2):24-27. |
| |
| Authors: | FAN Peng-pai SUI Li-fen CAO Yi-zhi |
| |
| Institution: | (Institute of Surveying and Mapping, Information Engineering University, Zheng Zhou 450052, China) |
| |
| Abstract: | Particle filter has been applied in many fields. As a recursive Bayesian estimation, particle filter uses a set of samples with weights to representing the conditional probability density function. The importance weights mix the states information and the observations, which will be influenced by outliers. A new approach, which is called robust particle filter, has been put forward based on the robust estimation. By use of the equivalent weight, this method can resist outliers. A numerical example contrast between this method and the traditional way is given. The result shows that the approach can control the influence by observation outliers effectively. |
| |
| Keywords: | particle filter robust estimation equivalent weight robust particle filter |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
