【作者】 宿德志；

【导师】 杨建坤；

【作者基本信息】 国防科学技术大学 ， 物理学， 2010， 硕士

【摘要】 本文可以分为两个部分,分别讨论了一种不依赖于光学镜头参数的全天星图识别方法和一种单星模拟器的调校方法。第一部分,一种改进的全天星图识别方法。改进了一种全天星图识别方法,此方法是利用观测星组成的三角形内角对于星敏感器相机的焦距f以及光轴偏差( x0 , y 0)等参数相对独立来进行星图识别,其识别特征为三角形内角。传统的星图识别方法,主要依据星对角距作为识别特征,其识别准确率依赖于星敏感器相机的焦距f以及光轴偏差( x0 , y 0)等参数的校正精度。而此方法根据导航三角形与观测三角形相似的特点,以三角形的内角作为识别特征,不依赖镜头参数,无需先验姿态信息。蒙特卡洛仿真结果表明,此方法的识别率约为93.37%;采用K矢量搜索法,识别时间约30ms;且在星点位置误差较大的情况下,仍然能较好的工作。第二部分,单星模拟器的调校方法。在传统的五棱镜法基础上,提出了一种单星模拟器出射光平行度的调校方法,本方法利用CCD相机取代了传统方法中的自准直仪,采用比较成熟的数字图像处理技术来提高调校的精度。理论分析表明,此调校方法的光平行度最大测量误差为0.49",可以满足高精度星模拟器的调校任务。实验结果表明,采用此方法的调校后,平行光发散角约为2.32",接近所选平行光管的衍射极限。另外,此调校方法配合电动导轨,MATLAB软件和Labview软件可以实现自动化快速调校。更多还原

【Abstract】 The approach of star identification and the method for adjusting of single star simulator is discussed.The first section is about the approach of star identification.Star identification is the most critical and important process for attitude estimation. The main purpose of the star identification process is to identify the observed stars with the corresponding cataloged stars. The precision of traditional star identification method highly depend on the calibrated accuracy of the star camera parameters, mainly the focal length f, and the optical axis offsets. When parameters of the camera are not accurate or when the camera is not well calibrated, the proposed star identification method becomes very suitable, because it does not require accurate knowledge of these parameters. Monte Carlo experiment shows that the probability of failing Star-ID is less than 6.63%. The adoption of the k-vector range search technique, makes this method very fast, the speed of this star identification method is less than 30ms.The second section is about the method for adjusting of single star simulator.Star simulator is necessary equipment for calibration and performance testing of star sensor in laboratory. A novel method for adjustment of star simulator has been proposed in this paper. This method is based on ordinary Penta prism, guide rail, Charge Coupled Devices (CCD) camera and sub-pixel centroid algorithm. The CCD camera is used to image the star light replicated by the Penta prism instead of the autocollimator compared with the traditional adjustment of collimator. The analysis shows that the limiting parallelism error of the single star simulator is less than 0.49″; the experiment shows that the error is less than 2.32″after adjustment using this proposed method. The difference between analysis and experiment is attributed to diffraction limit of the collimator. It can satisfy the high accuracy adjustment of single star simulator. Furthermore, with the software of MATLAB and LABVIEW, this method can work automatically.更多还原