混合线性/非线性模型的准高斯Rao-Blackwellized粒子滤波法  

Quasi-Gaussian Rao-Blackwellized Particle Filter for Mixed Linear/Nonlinear State Space Models

作　　者：庄泽森[1] 张建秋[1] 尹建君[1] 

机构地区：[1]复旦大学信息科学与工程学院,上海200433

出　　处：《航空学报》2008年第2期450-455,共6页Acta Aeronautica et Astronautica Sinica

基　　金：国家自然科学基金(60572023)

摘　　要：针对混合线性/非线性模型,提出一种新的递推估计滤波算法,称为准高斯Rao-Blackwellized粒子滤波器(Q-GRBPF)。算法采用Rao-Blackwellized思想,将线性状态与非线性状态进行分离,对非线性状态运用准高斯粒子滤波(Q-GPF)算法进行估计,并将其后验分布近似为单个高斯分布,再利用非线性状态的估计值对线性状态进行卡尔曼滤波(KF)估计。将Q-GRBPF应用于目标跟踪的仿真结果表明,与Rao-Blackwellized粒子滤波器(RBPF)相比,Q-GRBPF在保证估计精度的前提下有效降低了计算复杂度,计算时间约为RBPF的58%;与Q-GPF相比,x坐标与y坐标的估计精度分别提升了45%和30%,而计算时间也节省了约30%。A new recursive estimation algorithm, called the quasi-Gaussian Rao-Blackwellized particle filter （Q- GRBPF）, is proposed for filtering mixed linear/nonlinear state space models. The algorithm utilizes the idea of Rao-Blackwellized to separate the linear and nonlinear states. For the nonlinear states, the posterior distribu- tions of the estimates, which are achieved by the quasi-Gaussian particle filter （Q-GPF）, are approximated as Gaussian distributions. Also, the linear states are estimated by the Kalman filter（KF） with the estimated non- linear states. The simulation results of the proposed method applying to target tracking show that the proposed method only consumes 58% of the computing time required by the RBPF. Furthermore, compared with Q-GPF, the tracking accuracies of the proposed method for estimating x and y coordinate locations of the tracked target are respectively increased by 45% and 30% while 30% computing time is saved.

关 键 词：信号处理 准高斯Rao-Blackwellized粒子滤波器 仿真 混合线性/非线性 目标跟踪 

分 类 号：V249[航空宇航科学与技术—飞行器设计]