摘要
介绍了一种改进的粒子滤波(PF)算法——无味粒子滤波算法(UPF)。该算法结合UKF(unscentedKalman filter)和PF算法,利用UKF对非线性系统的处理能力,用UKF得到粒子滤波的重要性采样密度函数,从而克服了PF没有考虑最新量测信息和UKF只能应用于噪声为高斯分布的不足。在给出的闪烁噪声统计模型基础上,将UPF、PF算法在雷达目标跟踪中进行了比较,仿真结果表明该方法可以取得比标准的粒子滤波更快的滤波收敛性和更高的滤波精度。
Unscented particle filter(PF) is a new particle filtering method based on unscented Kalman filter and particle filter method, and can effectively cope with complicated nonlinear and non-Gaussian problems. The basic idea and algorithm description of unscented particle filter were presented. The UPF uses the unscented Kalman filter (UKF) to generate sophisticated proposal distributions. It can not only avoid the limitation of the UKF which only apply to Gaussian distributions but also avoid the limitation of the standard PF which can not include the new measurements. Then, the UPF and the PF were introduced to radar tracking based on the glint noise statistical model. The Monte Carlo simulation results show that the presented method has faster convergence rate and higher accuracy than the standard PF.
出处
《弹箭与制导学报》
CSCD
北大核心
2008年第1期79-82,共4页
Journal of Projectiles,Rockets,Missiles and Guidance
基金
国防预先研究基金资助
关键词
目标跟踪
粒子滤波
UKF
UPF
闪烁噪声
target tracking
particle filter
unscented Kalman filter
unscented particle filter
glint noise