TrackDrone data set - Siegendorf 2024

TrackDrone data set - Siegendorf 1

This data publication makes the first measurement campaign of the TrackDrone research project (FFG funded) publicly available. The campaign was conducted in May 2024 in Siegendorf, Austria together with Skyability GmbH. The project researches the trajectory estimation of measurement platforms, i.e. Uncrewed Aerial Vehicles (UAV) based on robotic total stations (RTS).

This data set consists of data acquired onboard the UAV (GNSS-based positions, laser scanner data and imagery), data acquired on the ground (position measurements of the UAV by two RTS) and reference data (terrestrial laser scanning point cloud and photogrammetric Ground Control Points). 

To give a brief overview: Two RTS were used to measure a reflector (Leica GRZ-122) mounted on a UAV; the UAV itself had a GNSS, an IMU, a laser scanner and a camera mounted. Thus, the multi-sensor system consists of six different sensors. To evaluate the data set, a terrestrial laser scanner (Riegl VZ-400i) was used to create a point cloud of the study area. All terrestrial measurement are connected by a geodetic network, which leads to local accuracies of less than 3 mm. In the study area, we also installed about 100 Ground Control Points as checkerboard targets. They are used to reference the photogrammetric data. 

The photogrammetric data is only provided as images with timestamps and no processing is published for this. The publication consists of only one UAV flight.

The data is available in raw and processed versions. The processed data considers the start and end of the flight to be airborne; the raw data includes static phases before and after the flight. 

The data set is analysed in two publications (https://doi.org/10.5194/isprs-annals-X-G-2025-213-2025) and one to be published at the DGPF Jahrestagung 2026 (Dammert et al., 2026 - https://dgpf.de/pub/tgb.html).

Content

Technical details

• The data formats are selected to be openly accessible; we recommend CloudCompare to view the point clouds
• The IMU data is given in digitial units and a scale factor to SI units must be empirically estimated, we estimated a factor of 1/86266.46 for the gyroscopes and 1/500000 for the accelerometers
• The lever arms w.r.t. the INS are  [in metres]:
  • RTS (prism) [-0.00683,0.014651,0.246592],
  • GNSS [-0.116,-0.009,-0.459],
  • LiDAR [0.1776,-0.0199,0.0004]
• A local East-North-Up coordinate system is used as project-based coordinate reference system. The included transformation data allows to transform the data in a global reference system, e.g. ETRS89 UTM33N
• The orientation unknown of the RTS data data is estimated to be zero
• Files with ending .pkl are pandas DataFrame objects and can be read using python package pandas (tested with version 3.0.0) with import statement pandas.read_pickle()
• Trajectories named *_sfi_* contain the positions estimated using a spatial forward intersection

Sensors used for data acquisition

• 2x Leica MS60 total stations
• RIEGL VUX-SYS laser scanner on the UAV
• Applanix AP-20X IMU on the UAV
• Sony Alpha7RII on the UAV
• Reference with RIEGL VZ-400i, Leica MS60 and SP80 & GS18 GNSS receivers

Further details

Please don't hesistate to contact us if anything is unclear about the data. The data displays a real environment, due to possible privacy concerns the data is published as restrcted record. This means if you want to use the data, we need to shortly confirm this.