迭代容积卡尔曼滤波算法及其应用

Iterated cubature Kalman filter and its application

将Gauss-Newton迭代和容积卡尔曼滤波(cubature Kalman filter,CKF)算法相结合,建立了一种迭代CKF(iterated CKF,ICKF)算法。该算法使用容积数值积分原则直接计算非线性随机函数的均值和方差,且在迭代过程中利用最新量测信息并改进迭代过程产生的新息方差和协方差,可获得较高的估计精度。针对弹道系数未知的再入弹道目标状态估计问题,仿真实验结果显示,该方法实现简单,比无迹卡尔曼滤波方法(unscentedKalman filter,UKF)及CKF方法效果要好。

An iterated cubature Kalman filter（ICKF） is proposed,which combines the Gauss-Newton iterate method with the cubature Kalman filter（CKF）.In the ICKF algorithm,cubature rule based numerical integration method is directly used to calculate the mean and covariance of the nonlinear random function,and the latest measurement,improved innovation covariance and cross-covariance are iteratively used in the measurement update,so the higher accuracy of state estimate is achieved.The ICKF algorithm is applied to state estimation for reentry ballistic target with unknown ballistic coefficient.The simulation results indicate that the implementation of the proposed method is easy and simple.Moreover,the higher accuracy of state estimation is obtained compared with UKF and CKF.