论文部分内容阅读
为了提高高阶容积卡尔曼滤波器(CKF)的滤波性能,提出一种基于矩阵对角化变换的高阶CKF算法.该算法基于高阶容积准则,利用矩阵对角化变换代替标准高阶CKF中的Cholesky分解,使得协方差矩阵分解后的平方根矩阵保留了原有的特征空间信息,状态统计量计算更加准确,从而提高了滤波精度;同时,矩阵对角化变换不要求协方差矩阵正定,增强了算法滤波稳定性.仿真结果表明,所提出的算法是可行而有效的,明显改善了标准高阶CKF的滤波效果.
In order to improve the filtering performance of high-order volume Kalman filter (CKF), a high-order CKF algorithm based on matrix diagonalization transformation is proposed.Based on high-order volumetric criterion, the algorithm uses matrix diagonalization transform instead of standard high-order CKF , The square root matrix after factorization of the covariance matrix preserves the original feature space information and the state statistic calculation is more accurate so as to improve the filtering precision. At the same time, the matrix diagonalization transformation does not require the positive covariance matrix to be positive, Which enhances the filtering stability of the algorithm.The simulation results show that the proposed algorithm is feasible and effective and obviously improves the filtering effect of standard high order CKF.