基于捷联惯导高保真误差模型的非线性滤波算法

Nonlinear Filter Based on High Fidelity SINS Error Model

针对捷联惯导系统在大姿态误差条件的对准问题,推导了高保真非线性误差模型。该模型采用欧拉角表示姿态误差,并用与导航解算相同的姿态更新算法准确描述了姿态误差的传播规律。利用该模型设计了扩展卡尔曼滤波器,并以二位置对准为例,比较该模型与线性模型和大方位误差模型的滤波效果。仿真实验结果表明在小角度误差和大方位误差条件下,基于高保真模型的滤波器估计精度都优于其它两种模型,且在大水平姿态误差条件下滤波精度也很高。

For the initial alignment of the strapdown inertial navigation system （SINS） under large atti- tude errors, a high fidelity model is deduced. The Euler angles are introduced to present the attitude errors. The propagation algorism same as the attitude update is applied to achieve accuracy error model. Based on the high fidelity model an extended Kalman filter （EKF） is designed. In two position alignment experi- ments, the performance of filters based on the high fidelity model is compared with those on linear and large heading uncertainty models. The simulation experiment results demonstrate that the filter based on the high fidelity model has more accuracy than that of any others filter and also has good performance in large tilt angle uncertainty condition.