改进的自适应卡尔曼滤波在SINS 初始对准中的应用

为提高 Sage-Husa 自适应滤波的稳定性，提出一种基于 UD 分解的改进自适应滤波算法。对在线估计的量测噪声协方差阵和状态估计误差方差阵采取UD分解的形式进行标示和更新，结合捷联惯导静基座初始对准模型，对改进自适应算法进行仿真测试。仿真结果表明：在先验量测噪声和状态估计协方差矩阵存在误差的情况下，新算法能够提高对准精度。

Application of Improved Adaptive Kalman Filter Algorithm in SINS Initial Alignment

In order to improve the stability of Sage-Husa adaptive filtering, an improved adaptive filtering algorithm that based on UD decomposition was proposed. The measurement noise covariance matrix that online estimated and state estimation error covariance matrix is updating by taking the form of UD decomposition and apply to the strapdown inertial navigation system of stationary base initial alignment model. The results show that when priori covariance matrix and state estimation error covariance matrix have an error, the new filter algorithm can improve the accuracy of the initial alignment.