顾及姿态线性误差的微小卫星面阵影像在轨几何检校

微小卫星由于平台体积、重量、能源等限制，其上搭载的姿态、位置测量设备精度不高，导致其直接对地定位误差较大。通过对某微小卫星嵩山地区的多景面阵影像进行姿态角常差检校，发现姿态角系统误差随时间线性变化的规律。为了提高定位精度，本文提出一种针对面阵的顾及姿态线性误差的偏置矩阵和二维探元指向角几何检校模型。相对于传统的姿态角常差检校模型，本文方法考虑了姿态角系统误差随时间线性变化的规律。试验结果表明，经过内外方位元素检校后，卫星的定位精度从数十千米提升到十米以内，相对于传统的常差模型，本文提出的检校模型有效地消除了姿态随时间线性变化的系统误差。

On-orbit geometry calibration of microsatellite area-array images based on attitude linearity error

The microsatellite has low accuracy of attitude and position measuring equipment,due to the limitations of platform volume,weight and energy,which results in large direct positioning error. Through the calibration of the constant angular error of the multi-view images in the Songshan area,we found that the system error of attitude angle varies linearly with time. This paper proposes an on-orbit geometry calibration model based on two-dimensional direction angle of detector and offset matrix with attitude linearity error to improve the positioning accuracy. Compared with the traditional constant angular model,this paper considers the law of linear variation of attitude angle system with time. The experimental results show that after the internal and external orientation elements are calibrated,the positioning accuracy of the satellite is increased from tens of kilometers to less than ten meters. The proposed calibration model effectively eliminates the system error of attitude angle varying linearly with time compared with the traditional constant angular model.