一种基于模式匹配的自主星图识别算法
An Autonomous Star Identification Algorithm Based on Patterns Matching
-
摘要: 为便于星光定姿系统在线应用,提出了一种新的基于模式匹配的自主星图识别算法. 该算法首先建立了一种新的星模式描述方法,然后用该模式描述方法建立观测星的模式和导航星模式数据库,最后通过观测星模式与导航星模式数据库进行匹配,完成星图识别. 试验结果表明,该星图识别算法与传统的算法相比,对观测位置误差具有更好的适应能力,需要的存储空间更少,计算速度更快.Abstract: In order to facilitate the online application of stellar attitude determination system, a new autonomous star identification algorithm based on star pattern matching was presented in this paper. First, a new approach was proposed for describing the patterns of star image, and then the observed star pattern and the database of navigation stars patterns were constituted by using the star pattern description. Finally, the star recognition was conducted by comparing the observed star pattern vectors with an onboard database pattern vectors. Experiment results show that the proposed algorithm is very robust to the noise of observed star position. Also, it uses less computer resources and achieves faster calculating speed compared with the conventional algorithm.
下载: