Kriging方法结合最小二乘配置在GPS高程拟合中的应用  

Combination of Kriging and least squares collocation and its application in GPS height fitting

作　　者：李军海[1,2] 文汉江[1] 方爱平[1] 刘焕玲[1,2] 

机构地区：[1]中国测绘科学研究院,北京100830 [2]辽宁工程技术大学测绘与地理科学学院,辽宁阜新123000

出　　处：《测绘科学》2011年第1期99-101,共3页Science of Surveying and Mapping

基　　金：国家自然科学基金(40874012);武汉大学地球空间环境与大地测量教育部重点实验室开放研究基金(07-05)

摘　　要：本文利用Kriging方法结合最小二乘配置将GPS高程转换成正常高。研究了将Kriging方法中的变异函数用于计算最小二乘配置中的协方差的方法,并对一局部GPS水准网的高程作了拟合计算。通过将最小二乘配置法与平面拟合模型和多面函数拟合模型等进行比较,其外符合精度从最大的±0.0277m提高到±0.0162m。Based on least squares collocation, a conversion method of GPS geodetic height to normal height using Kriging algorithm was proposed in this paper. It is studied that the variogram in the Kriging was used to calculate the covariance in the least squares collocation, and a local GPS leveling network height example was analyzed. By comparing the results from least squares collocation with that from the model of plane fitting and polyhedral function fitting, it was found that the accuracy of external coincidence improved from±0. 0277m to ±0. 0162 m.

关 键 词：最小二乘配置 变异函数 协方差函数 高程拟合 

分 类 号：P228[天文地球—大地测量学与测量工程]