论文部分内容阅读
高斯粒子滤波算法重要性权值方差不会随迭代次数的增加而增加,能够较好地解决粒子退化问题,但其重要性密度函数没有考虑最新的量测信息,导致有效粒子数减少,算法滤波性能下降.针对该问题,提出一种基于Gaussian-Hermite滤波(GHF)的高斯粒子滤波算法,采用GHF构造高斯粒子滤波的重要性密度函数,考虑最新的量测信息,增加有效粒子数,提高算法的滤波精度.仿真结果表明,所提出算法的滤波精度明显优于高斯粒子滤波算法.
The weight of Gaussian particle filter algorithm does not increase with the increase of iteration number, which can solve the problem of particle degeneration. However, the importance density function does not consider the latest measurement information, which leads to the reduction of effective particle number. This paper proposes a Gaussian particle filter algorithm based on Gaussian-Hermite filter (GHF), using GHF to construct the importance density function of Gaussian particle filter, taking into account the latest measurement information, increasing the number of effective particles and improving the algorithm The simulation results show that the proposed algorithm has better filtering accuracy than Gaussian particle filter.