The study on the synthesis approximate algorithm of GPS height conversion

Abstract

Studying on the GPS height transformation algorithm to improve the accuracy of GPS height conversion has always been a hotspot in the field of geodesy. At present, there are many methods of converses GPS height into normal height, the most common of which is the method of numerical approximation, and its algorithm is mostly confined to how to choose a suitable function model or statistical model to numerical approximation. Aiming at the singularity of GPS height conversion methods, this article presents a comprehensive approximation algorithm, which combine the functional approximation models with statistical approximation models in order to take full advantage of the regularity of function approximation model and the flexibility of the statistical approximation model. Based on the analysis of the actual engineering data, the results show that the accuracy of GPS height transformation based on the algorithm of integrated approximation approach is superior to the single function transformation model or a single statistical model. At the same time, it also relaxes the selection requirements of function model and statistical model.