摘要
用迭代扩展卡尔曼滤波方法来生成辅助粒子滤波的重要性密度函数,得到了一种新的改进的滤波算法:迭代扩展卡尔曼辅助粒子滤波.仿真结果表明,该算法在观测量较精确的情况下改善了粒子权值分布,减轻了粒子退化现象,该算法不仅要优于已有的滤波方法,而且比无忌卡尔曼粒子滤波运行时间短.分析了各算法改进的原因及适应情况.
In this paper, an improved filter procedure, named the iterated extended Kalman auxiliary particle filter （APF-IEKF）, is proposed. The proposed algorithm consists of an auxiliary particle filter that uses an iterated extended Kalman filter to generate the importance proposal distribution. The simulation results show that the new procedure improves the distribution of the weights in the case of accurate measurement, which mitigates the effects of particle degeneracy problem. The experimental results also illustrate that the improved particle filter is superior to the existing filters and that it has less running time than the unscented Kalman particle filter（PF-UKF）. Additionally, the performance of these algorithms is compared and some reasons of performance improving of each algorithm are analyzed.
出处
《系统工程学报》
CSCD
北大核心
2012年第5期593-599,共7页
Journal of Systems Engineering
基金
国家自科基金委创新群体资助项目(70921001)
湖南省科技厅国际合作重点资助项目(2009WK2009)
关键词
粒子滤波
辅助粒子滤波
迭代卡尔曼滤波
重要性密度函数
particle filter
auxiliary particle filter
iterated extended Kalman filter
importance density function
