无陀螺卫星姿态的二阶插值非线性滤波估计

Second-order nonlinear interpolation filtering for gyroless satellite attitude estimation

采用四元数作为姿态描述参数 ,提出了一种确定无陀螺卫星姿态的新方法 ,即二阶插值非线性姿态估计算法 ,它能够利用星敏感器提供的矢量观测信息准确地估计三轴稳定卫星的姿态。这种姿态估计算法的实现非常简单 ,其运算量与传统的扩展卡尔曼滤波姿态估计算法相当 ,但滤波性能却与基于二阶泰勒级数近似得到的非线性姿态估计算法一致。而且 ,在二阶插值非线性姿态估计算法中 ,不会遇到由四元数正交约束所造成的协方差阵奇异性问题。

A second order nonlinear interpolation filter is presented for gyroless attitude estimation.The filter can accurately estimate the quaternion that describes the attitude of a three axis stabilized satellite using only star sensor measurements.The implementation of the new attitude estimation algorithm is significantly simple.The computational burden of the new algorithm is comparable to that of the conventional extended Kalman attitude estimation algorithm.But the performance of the former is superior to that of the latter.The attitude estimation accuracy of the new algorithm is the same as that of the nonlinear attitude estimation algorithm based on second order Taylor series approximations.However the operation efficiency of the new algorithm is evidently higher than that of the Taylor series based nonlinear attitude estimator.In addition,no derivatives are required in the new algorithm.Therefore,the singularity of the covariance matrix caused by the constraint on the attitude quaternion normalization is avoided in the second order nonlinear interpolation attitude estimation algorithm.The good performance of the new attitude estimtor is demonstrated by simulation test.