Key facts

• Fields of application:
  railway positioning, sensor fusion, sensor models
• Available data:
  2x GNSS, 1x IMU
• Rail-track characteristics:
  ≈120 km on conventional and secondary line (tight curves, steep slopes, forested embankment)
• Available reference data:
  Open GNSS/IMU EKF-fusion solution (loosely coupled), high precision 3D track-map
• Structure:
  This data set follows the data sharing principles of the LRT (localization reference train) initiative that are available at lrt-initiative.org.

About

This data set can be used for rail vehicle positioning experiments. It contains measurements of an 6-DOF IMU and two GNSS receivers. The senors were mounted on a regular rail vehicle during a trip from Chemnitz (Germany, Saxony) to Schwarzenberg (Germany, Saxony) and back.

The recorded data have been pre-processed to have common time and coordinate frames. Furthermore, a simple loosely coupled GNSS/IMU positioning solution is provided which can be used as a baseline for more advanced fusion approaches.

All MATLAB scripts used to process the raw data and to calculate the GNSS/IMU positioning solution are provided within the data set.

The data can be used as a starting point for own work. Special features of this data set are its covered terrain (tight curves, steep slopes, forest embankment), its test-track (often used as test-track for new railway equipment) and the availability of a high precision 3D track-map.

Similar data sets

• https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/2292.2
• https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/2529

Scripts

• https://github.com/tud-rmr/railway_dataset_c_20181024

BibTex

@Misc{WinterRailDataSetOctober2018,
title = {Rail Vehicle Positioning Data Set: Lucy, October 2018},
author = {Winter, Hanno},
doi = {10.25534/tudatalib-360},
publisher = {Technische Universität Darmstadt},
year = {2020},
}