迭代自适应容积卡尔曼滤波算法

Iterative adaptive volume Kalman filter algorithm

针对标准容积卡尔曼滤波(CKF)算法跟踪精度低、稳定性差的问题,提出了一种采用优化迭代测量更新过程方法,并将其引入到自适应的容积卡尔曼滤波算法中。该算法不仅保证了迭代算法的有效性,还在很大程度上提高了CKF算法的精度、增强算法的稳定性,新算法还具有应对噪声统计特性变化的自适应能力。采用非线性高斯模型进行仿真实验与分析,实验中给出了扩展卡尔曼滤波(EKF)、无际卡尔曼滤波(UKF)、CKF、以及改良的迭代自适应容积卡尔曼滤波(IDCKF)等算法的目标跟踪滤波估计结果。并根据均方根误差、对目标跟踪位置与速度的均方根误差以及执行时间来证明新算法的有效性。

Aiming at the problem of low tracking accuracy and poor stability of standard volume Kalman filter(CKF)algorithm,a method of optimizing iterative measurement update process is proposed,and it is introduced into adaptive volume Kalman filter algorithm.This algorithm not only guarantees the effectiveness of the iterative algorithm,but also greatly improves the accuracy and stability of CKF algorithm.The new algorithm also has the adaptive ability to deal with the change of noise statistical characteristics.A non-linear Gauss model is used to simulate and analyze the target tracking filter.Experiments show the estimation results of extended Kalman filter(EKF),infinite Kalman filter(UKF),CKF,and improved iterative adaptive volume Kalman filter(IDCKF).The effectiveness of the new algorithm is proved by the root mean square error,the root mean square error of tracking position and velocity and the execution time.