BDS/GPS组合观测值受观测环境的影响难免存在粗差或多路径等误差,基于最小二乘的方差分量估计不具有抵抗粗差或多路径等误差的能力,严重影响了随机模型的有效性. 针对这一问题,基于抗差估计思想,引入双因子等价权函数,给出了抗差方差分量估计的实现过程. 通过对BDS/GPS短基线实测数据进行分析和计算,表明抗差方差分量估计可有效抵抗粗差或多路径等误差,提高了随机模型的可靠性,抗差估计成果具有更高的内符合精度.%Variance component estimation is often used to determine the weighs of the observation values in com-bined BDS/GPS positioning. The observations often have gross errors or large errors which have a serious impact on the effectiveness of stochastic model and distort the positioning result because of the variance component estimation based on least squares can't resist gross error. To resolve this problem,a method which combines variance component estimation with the thoughts of robust estimation and implementation peocedures are introduced. The testing data indicates that ro-bust variance component estimation can largely improve the robustness and reliability of combined BDS/GPS positioning, and the results of combined positioning have a higher accuracy and more practical value.
展开▼
