稳健估计在GNSS高程拟合中的粗差探测

摘要

由于最小二乘法不能有效地抵抗粗差,而控制点的平面坐标和高程异常值中不可避免地含有误差,对应用最小二乘法和稳健估计法在GNSS高程拟合中的粗差探测进行探讨.通过对不同数量控制点的高程异常观测值中加入粗差,采用两种算法在求解GNSS高程拟合中的精度进行分析比较,并对粗差在稳健估计中的干扰范围进行研究,结果表明,稳健估计具有抵抗多个粗差的能力.%The method of least-squares cannot resist gross of errors effectively.There are errors inevitably in the plane coordinate and height anomaly of the control points.This article tries to discuss the solution about detection of gross-error in GNSS elevation fitting with method of least-squares and robust estimation.The two algorithms are used to analyze and compare the precision in the solution of GNSS elevation fitting,and it studies the disturbance range of gross error in robust estimation by adding gross error to observations of elevation anomaly in different number of control points.The result of the study indicates that robust estimation has the ability of resisting multiple gross errors.