GPS相关观测量的粗差探测方法比较及实例分析

Comparison of Gross Error Detection Methods in GPS Correlation Observation and Case Analysis

在GPS网平差中,其观测值为三维基线向量,基线向量三个分量间存在较强的相关性,三个分量即相关观测量。普通最小二乘法是一种基于独立观测值的稳健估计法,对独立观测值的粗差探测具有较好效果,但却无法兼顾观测值之间的相关性,以至于不能准确发现粗差,因此本文通过在传统稳健估计的基础上,在定权时充分考虑相关观测量之间相关性的不变性,不断构造严格对称的方差-协方差阵,并对其进行不断扩大,并通过VB进行编程及实例比较分析,从而论证其具有准确定位相关观测量中的粗差所在位置的能力,并在进行有效平差后精度高的效果。

In the adjustment of GPS network,the observation value is three-dimensional baseline vector,and there is strong correlation among three components of baseline vector,which are correlation observation.The general least square method is a robust estimation method based on independent observations,which has a good effect on the detection of gross errors of independent observations,but it can’t take into account the correlation between observations,so it can’t find gross errors accurately.Therefore,based on the traditional robust estimation,this paper fully considers the invariance of correlation between related observations in the determination of weights,and constantly constructs strict The variance covariance matrix of lattice symmetry is expanded continuously,and through VB programming and comparative analysis of examples,it is proved that it has the ability to accurately locate the position of the gross error in the relevant observation value,and the effect of high precision after effective adjustment.