摘要
为解决粒子滤波(PF)固有的退化现象及因简单重采样引起的粒子匮乏问题,采用扩展卡尔曼滤波(EKF)来优选PF的重要性分布,并对重采样方法进行改进。通过理论分析及针对全球定位系统(GPS)的计算机仿真,对比扩展卡尔曼滤波(EKF)、扩展卡尔曼粒子滤波(EKPF)以及改进的EKPF算法来实现导航定位的定位估计精度与效率,分析在不同条件状况下的最佳非线性滤波算法。实验结果表明,与其它方法相比,该算法在高动态、高机动状态下性能得到了明显的改善。
To solve the degeneracy phenomenon of Particle filter（PF） and sample impoverishment problem caused by simple random resampling,based on the extended Kalman particle filter（EKPF） which selects the importance distribution of PF by the extended Kalman filter（EKF）,we improved the resampling method for the EKPF.According to theoretical analysis and computer simulation in positioning and navigation in the global positioning system（GPS）,we compared the positioning accuracy and efficiency from EKF,EKPF and the improved EKPF to give an evaluation for these algorithms in different statuses.The simulation results demonstrate that the performance of the proposed method has obvious improvement compared to other methods at the status of high dynamic,high mobility.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第11期2523-2526,共4页
Computer Engineering and Design
关键词
非线性滤波
扩展卡尔曼滤波
粒子滤波
重采样
全球定位系统
nonlinear filtering
extended Kalman filter
particle filter
resampling
global positioning system
