基于自适应扩展Kalman滤波的SINS/GPS深组合研究  

Study on SINS/GPS Deeply Integrated Navigation Based on Adaptive Extended Kalman Filter

作　　者：张品秀[1] 黄操军[1] 乔相伟[2] 

机构地区：[1]黑龙江八一农垦大学信息技术学院,黑龙江大庆163319 [2]哈尔滨工程大学自动化学院,哈尔滨150001

出　　处：《传感技术学报》2010年第3期408-412,共5页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金资助(60704018)

摘　　要：针对SINS/GPS组合导航系统噪声随时间变化引起卡尔曼滤波精度下降的问题,提出了一种噪声统计特性在线估计的自适应扩展卡尔曼滤波算法。算法首先基于新息序列实现了对观测噪声协方差的实时估计,然后基于系统方程采用协方差匹配算法完成了对过程噪声的实时跟踪。算法中尺度因子的引入进一步减小了泰勒展开造成的高阶截断误差,提高了滤波精度。仿真实验结果说明,与传统卡尔曼滤波算法相比,该算法能够实现对过程和观测噪声的完全估计,鲁棒性和精度都有明显提高。To avoid the precision declination of Kalman filtering caused by the noise variation, an adaptive extended Kalman filtering is proposed to estimate noise statistical property on line in SINS/GPS integrated navigation system. First, measurement noise covariance is estimated through innovation sequence online, then the covariance matching algorithm is used to track the process noise real-time based on the system equation. Additionally, scale factor is introduced to reduce truncation error caused by Taylor formulation and thus improve estimation accuracy. The Simulations results show that, compared with the traditional Kalman filtering algorithm, the proposed algorithm is able to estimate the changes of both process and observation noise statistics simultaneous, and have higher precision and more robustness.

关 键 词：UKF AEKF SINS/GPS 噪声在线估计 协方差匹配 

分 类 号：U666.11[交通运输工程—船舶及航道工程]