GPS高程拟合的不同估计方法比较  

Comparison of various estimation methods in GPS elevation fitting

作　　者：唐诗华[1,2] 袁隆疆 王江波 肖阳 蒲伦 TANG Shi-hua;YUAN Long-jiang;WANG Jiang-bo;XIAO Yang;PU Lun(Guangxi Key Laboratory of Spatial Information and Geomatics,Guilin University of Technology,Guilin 541004,China;College of Geomatics and Geoinformation,Guilin University of Technology,Guilin 541004,China)

机构地区：[1]桂林理工大学广西空间信息与测绘重点实验室,广西桂林541004 [2]桂林理工大学测绘地理信息学院,广西桂林541004

出　　处：《桂林理工大学学报》2018年第3期501-506,共6页Journal of Guilin University of Technology

基　　金：广西空间信息与测绘重点实验室主任基金项目(16-380-25-13);广西空间信息与测绘重点实验室项目(15-140-07-05)

摘　　要：通过GPS高程异常平面拟合实例,系统地介绍了最小二乘估计、总体最小二乘估计、稳健最小二乘估计、稳健总体最小二乘估计、半参数估计等5种估计方法的原理和特点,比较了它们的识别模型参数、减弱系统误差、抵御粗差和不同类型误差的能力。结果表明:总体最小二乘估计在GPS高程拟合中稳定性较弱,IGGⅢ抗差方案会放大随机模型误差对平面拟合模型的影响,半参数估计只有在系统误差存在下与最小二乘估计参数值相差大,最小二乘估计在高程拟合中精度最高。The principles and characteristics of different estimation models which including least squares,total least squares,robust least squares,robust overall least squares and semi-parametric model are introduced to compare with the error value characteristics of GPS height.And robust overall least squares model is emphasized.The abilities of identifying model parameters,the attenuate systematic errors,resisting the gross error and different types are compared with these estimation models.The error value characteristics of GPS height results show that as the stability of the total least squares estimate is poor,the IGGⅢanti-difference scheme will magnify the error of random model on plane fitting model.There will be great difference between the semi-parametric model and the least squares estimate model in the presence of systematic errors.And the least squares estimate model has the highest fitting precision.

关 键 词：稳健总体最小二乘估计 半参数估计 GPS高程异常拟合 适用范围 最优估计 

分 类 号：P228.9[天文地球—大地测量学与测量工程]