改进的UKF在惯导平台误差模型辨识中的应用  

Application of improved UKF in error model identification of inertial navigation platform

作　　者：柳明[1] 刘雨[1] 苏宝库[1] 

机构地区：[1]哈尔滨工业大学空间控制与惯性技术研究中心,哈尔滨150001

出　　处：《控制与决策》2009年第1期129-132,136,共5页Control and Decision

基　　金：国家安全重大基础研究项目(973-61334)

摘　　要：为减小建模误差,建立了基于直接法进行惯导平台误差模型辨识的非线性模型.Unscented Kalman滤波(UKF)是一种新的非线性滤波算法,为此将其引入惯导平台的误差模型辨识中.针对系统模型的特点,对标准UKF算法进行了简化改进.改进的UKF算法计算量小、结构简单,滤波精度与标准UKF一致.同时应用扩展Kalman滤波(EKF)算法和改进的UKF算法进行了惯导平台误差模型辨识仿真研究.仿真结果表明,与EKF算法相比,改进的UKF算法的滤波精度显著提高.To reduce the modeling error, the nonlinear model of direct method based error model identification of inertial navigation platform is given. The Unscented Kalman filter（UKF） is a new nonlinear filtering algorithm. The UKF algorithm is introduced to the error model identification of inertial navigation platform. According to the peculiarity of the system model, the UKF algorithm is improved. The improved algorithm has the merits of higher calculation speed and simpler configuration, and its precision is identical to the UKF algorithm. The improved UKF algorithm and the extended Kalman filter （EKF） are used to the error model identification of the inertial navigation platform. The simulation results show that, compared with the EKF algorithm, the improved UKF algorithm can enhance the filtering precision.

关 键 词：惯导平台 非线性滤波 UKF算法 模型辨识 

分 类 号：TN911.72[电子电信—通信与信息系统]