CHANGE POINT DETECTION METHOD OF RTK-GNSS DATA CONSIDERING MEASUREMENT ERROR IN SLOPE MANAGEMENT

Abstract

 In recent years, various monitoring data has been obtained that can be used for slope disaster prevention. A change point detection method using ChangeFinder has been proposed based on RTK-GNSS, a satellite-based positioning system, but its application remains to be expanded for practical use. In this paper, we proposed a method for processing measurement errors in GNSS data and attempted to improve detection accuracy. Specifically, we propose a sidereal time-based correcting method to process multipath errors and a coincidence error processing model to distinguish the true position from GNSS data containing coincidence errors, and perform change point detection with the help of sequential estimation using particle filters. Furthermore, we validated the results using GNSS data observed on an actual expressway and showed that change points can be detected when the displacement of the slope exceeds 1.5 times the sample standard deviation of GNSS data.