选权拟合的粗差参数化平差模型

Parameterized Gross Error Adjustment Model Based on the Fitting Method by Selection of the Parameter Weight

由于数据探测法、多维粗差同时定位与定值法、拟准检定法以及部分最小二乘法不能同时解算待估参数和粗差,提出了基于选权拟合的粗差参数化平差模型。首先推导了选权拟合的等价模型;在此基础上提出了参数化粗差的平差模型。由于粗差参数化之后观测系统呈现出不适定性,因此,对非粗差部分采取平方和最小的约束,实际上是构造虚拟观测方程。但是,由于事先并不知道粗差的位置,提出了通过搜索来构造虚拟观测方程系数矩阵的方法。算例表明,这种方法用于GPS单点定位可以有效地消除粗差观测的影响。

Due to the fact that the parameters to be estimated and gross errors cannot be solved simultaneously by data-snooping method,LEGE,QUAD and PLS,the parametric gross error adjustment model is proposed based on the Fitting Method by Selection of the Parameter Weight.Firstly,the equivalent model of the Fitting Method by Selection of the Parameter Weight is deduced.Based on this,the adjustment model of parameterized gross error is put forward.Since the observation system is ill-posed after the parameterization of gross error,the constraint of the minimum sum of squares on the non-gross error part is employed,which is actually the construction of virtual observation equation.However,because the position of gross error is not known in advance,a method of construct the coefficient matrix of virtual observation equation by searching is proposed.The example shows that this method can remove the effect of gross error observation effectively when used for GPS single point positioning.