多视场星敏感器结构参数标定方法

提出一种基于星间角距不变原理的多视场(FOV)星敏感器结构参数标定方法.这种方法以欧拉角表征多视场星敏感器各个子视场之间的旋转关系,利用识别得到的各个视场星点的坐标信息和赤经赤纬信息,计算出多对星光矢量来建立标定模型和目标函数,然后使用L-M算法优化目标函数并解算出各个子视场之间的结构参数.此方法不需要外部姿态测量仪器辅助,可用于在轨和地面标定.在全天球随机抽取多个姿态生成多视场的仿真星图用于标定,并采用星内角统计偏差作为结构参数标定精度的评价指标.这种方法能够准确求解多视场星敏感器的结构参数.星内角统计偏差的平均值在星点位置噪声标准差为0.1像素的仿真试验中为1.3",在外场观星的实际试验中为6.4". 

Structural parameter calibration method for multiple field of view star tracker

Based on rotation-invariant star-pair angular distances, a calibration algorithm was proposed to calibrate structural parameters of multiple field of view (FOV) star tracker. With the rotation relationships of each subfield of view were described by three Euler angels, the calibration scheme and objective function were generated by some starlight vectors of star pairs, which were obtained from the image coordinates and the corresponding celestial coordinates of observed star points. Levenberg-Marquardt (L-M) algorithm optimized the objective function and provided the relative Euler angles of each sub FOV. External attitude-measurement devices are not necessary, and the method can be used for on-obit and ground calibrations. Several attitudes were randomly selected to create star images used for calibration in the celestial sphere, then the inter star-angle statistical residual was chosen as an evaluation standard for accuracy calibration. The results show that this method is effective to calculate the accurate structural parameters. Under the positional noise level of 0.1 pixel, the inter star-angle statistical residual can reach 1.3 arc sec on average, and it can reach 6.4 arc sec on average when the method is used in outdoor experiments.