EKF、UKF、PF目标跟踪性能的比较

雷达系统的非线性目标跟踪已被人们广泛重视。扩展卡尔曼滤波器（EKF）是将卡尔曼滤波器（KF）局部线性化，其算法简单、计算量小，适用于弱非线性、高斯环境下。不敏卡尔曼滤波器（UKF）是用一系列确定样本来逼近状态的后验概率密度，在高斯环境中，对任何非线性系统都有较好的跟踪性能。粒子滤波器（PF）是用随机样本来近似状态后验概率密度函数，适用于任何非线性非高斯系统。文中通过仿真实验，对三者的性能进行了仿真比较，结果证明在复杂的非高斯非线性环境中，粒子滤波器的性能明显优于另外两种滤波器，但计算复杂，耗时长。

Comparing of Target-Tracking Performances of EKF,UKF and PF

Nonlinear target-tracking methods have been widely researched and used in radar system. Extended Kalman filter(EKF) based on local linearization of KF, is easy to realize and has good performance in Gaussian and mild nonlinear environment. Unscented KF(UKF) utilizes a set of definite samplings to approximate posterior probability density function, while particle filter(PF) uses random particles. Hence, UKF is suitable for any nonlinear environment but Gaussian environment, but PF plays good act in any nonlinear and non-Gaussian environment. By simulation experiments, their performances are compared. The results prove the tracking performance of PF is much better than ones of EKF and UKF in complex environment, but computation of PF is much larger than ones of EKF and UKF.