一种新的自适应粒子滤波算法*

在对实时性要求高的系统中,粒子数的选择对粒子滤波至关重要,需要在滤波精度与计算复杂度乏间取折衷.提出了一种基于决策规则的自适应粒子滤波算法,在定义综合性能代价函数的基础上,推导出粒子数与滤波误差方差之间的关系,使得在粒子滤波过程中,可以根据实际滤波情况在线调节粒子数,以使滤波的综合性能达到最优.通过Monte Carlo仿真实验表明,新算法在与基本粒子滤波算法(PF)保持同等精度的条件下,大大降低了算法的计算量,约为PF的六分之一.

New Adaptive Particle Filtering Algorithm

For the system which has the exigent request for real time,the choice of sampling numbers is extremely important for particle filtering and it needs to make a compromise between filtering precision and compute complexity.This article proposes an adaptive particle filtering algorithm based on decision-making rule,and on the basis of defining the integrate performance cost function(IPC),the relationship between the number of particles and filtering error variance is educed so that the number of used particles...