模糊度二维搜索的姿态测量改进算法

值域算法由于采用遍历姿态角的方式建立搜索空间,从而会造成初始化时间较长的问题,故提出了一种基于窄巷组合的模糊度二维搜索算法:首先基于卫星空间几何关系对参考星与两颗主星间的窄巷主模糊度范围进行约束,利用其与姿态角之间的关系推得全体候选姿态角,从而以一种非遍历且不遗漏正确候选姿态角的途径得到搜索范围。然后基于值域二维搜索模型求解各候选姿态角相对应的窄巷模糊度浮点解,就近取整后依次采用模糊度整数特性法、基线先验信息法以及基线残差平方和最小准则固定整周模糊度。试验结果表明:文中算法相比原算法,不仅初始化时间由先前的153 s缩短至76 s,而且姿态精度总体提高近15%。

An Improved Attitude Measurement AlgorithmBased on Planar Search of Ambiguity

Aimed at the problem that because the result zone algorithm is in search of space by adopting the method of traversing all attitude angles, the initialization time is comparative long,an improved planar ambiguity searching method is proposed based on the combination of narrow lanes. First, based on the relationship of satellite space geometry to restrain the narrow lane ambiguity scale between reference star and two main stars, all the candidate attitude angles can be obtained by using the relationship between the attitude angle and the narrow ambiguity, thus obtaining the searching range in a way neither traversing nor missing the correct candidate. And then, each corresponding narrow-lane ambiguity float solution is derived from result zone planar searching model. and the narrow-lane ambiguity is fixed by using ambiguity integer characteristics method, baseline length constraint method and least square sum of baseline residual. The experiment results show that the new method can effectively reduce the initialization time from 153 s to 76 s and the attitude accuracy invreases nearly by 15% comparing with the original algorithm.