迭代容积粒子滤波算法在SINS初始对准中的应用

摘要

在大方位失准角误差的条件下,捷联惯导系统(SINS)初始对准误差模型是非线性的,可以采用粒子滤波(PF)方法进行处理.针对标准PF算法中存在的重要性密度函数难以选取的问题,提出了一种新的迭代容积粒子滤波(ICPF)算法.将Gauss-Newton迭代和容积卡尔曼滤波(CKF)算法相结合,得到迭代CKF(ICKF)算法.该算法利用最新量测信息改进迭代过程中产生的新息方差和协方差,可获得较高的估计精度.由ICKF算法获得粒子滤波算法的重要性密度函数,有效地抑制了粒子退化现象.SINS大方位失准角初始对准的仿真结果和实验结果表明:该算法的滤波精度高于标准PF算法和容积PF(CPF)算法,是一种非常有效的非线性滤波算法.%The error model of strap-down inertial navigation system (SINS)initial alignment is nonlinear under large azimuth misalignment angle condition,and it can be processed using particle filtering (PF)algorithm.Aiming at the problem that importance density function of standard particle filtering algorithm is difficult to select,a new algorithm of iterated cubature particle filtering (ICPF) is proposed.Gauss-Newton iteration and cubature Kalman filtering(CKF) algorithm are combined to obtain iterated CKF(ICKF) algorithm.The latest measurement information are used in this algorithm to improve the innovation variance and covariance in iterative process,and obtain higher estimation precision.The importance density function of PF is obtained by ICKF algorithm and particle degradation phenomenon is effectively restrained.Simulation and experimental results show that,the filtering precision of this algorithm is higher than standard PF and CPF algorithm and it is a very effective nonlinear filtering algorithm.