论文部分内容阅读
为解决传统粒子滤波算法中影响状态估计性能的采样枯竭问题,提出一种高斯混合粒子滤波(GMPF)算法,基于Sigma点卡尔曼滤波(SPKF)和粒子滤波的特点,采用加权EM算法取代传统粒子滤波的再采样过程,减弱了采样枯竭的影响,增强了算法的估计性能.对捷联惯导系统静基座大方位失准角初始对准的仿真结果表明,该算法的估计精度优于扩展卡尔曼滤波.
In order to solve the problem of sample depletion affecting the state estimation performance of traditional particle filter, a Gaussian mixture particle filter (GMPF) algorithm is proposed. Based on the characteristics of Sigma point Kalman filter (SPKF) and particle filter, a weighted EM algorithm is used to replace the traditional particle The resampling process reduces the influence of sample exhaustion and enhances the estimation performance of the algorithm.The simulation results of the initial alignment for the large azimuth misalignment angle of the strapdown inertial navigation system static base demonstrate that the proposed algorithm is superior to the extended Kalman filter.