GNSS Data Driven Clustering Method for Railway Environment Scenarios Classification

This paper introduces a clustering method based on recorded GNSS historical NMEA data collected by GNSS receiver, which mounted on the roof of the locomotive. The method aims at establishing the environment scenarios categories according to the skyplot along the railway track. Azimuth, elevation angle and signal-to-noise ratio (SNR) in the GNSS receiver NMEA data is used to generate environment features, then put into the clustering algorithm: dividing azimuth into 6 sections, the features are generated according to the elevation angle and SNR of the visible satellites in each section. Analyzing cluster result with the SNR, visible satellite number and horizontal dilution of precision (HDOP), the environment scenarios are distinguished through the proposed algorithm in this paper. The results show that algorithm can accurately regenerate the skyplot in different railway environment scenarios, which is helpful for environment scenario prediction and error estimation for unknown environments.