GPS精密测量中系统误差的分离方法

Separation of Systematical Errors from GPS Precise Measurements

将系统误差作为待估参数纳入观测方程。观测方程属于观测数小于待估参数个数的秩亏方程。本文提出一种新算法。通过降维变换 ,并借鉴拟稳平差思想 ,采用“选群拟合”附加约束 ,可求得基线分量等确定性参数的估值以及系统误差估值 ,削弱了系统误差对 GPS精密定位结果的影响。用两个 GPS基线实测算例 ,比较了新方法与以往算法的效果。最后讨论了新方法在

Systematical errors are considered as the estimated parameters which bring into the GPS observation equations. In this case, the observation equations belong to the rank deficient ones in which the number of the observations is less than that of the parameters. To resolve this type problem, a new algorithm is presented. Firstly, the dimension of the parameters is decreased by a transformation. Then the idea of 'the quasi stable adjustment' is used for reference, and 'the selected group fitting' is taken in the scheme. Finally, the estimators of the determinated parameters, such as the components of a baseline, and systematical errors can be found out by adding the conditions for minimum of the normal of 'the quasi accurate parameters'. By this way, the effects of the systematical errors on the results of GPS precise position is obviously declined. Two examples of GPS baseline are listed to compare the result of the new method with that of the classical one. At last, the appropriate conditions using the new method in GPS surveying are discussed.