TrackDrone data set - Siegendorf 2026

TrackDrone data set II 

This data set is related to TrackDrone data set - Siegendorf 2024 (10.48436/yzk5x-49910).

However, it is an independent data set that improves the existing data set. The main improvement is the availability of synchronised image data from the image-assisted total stations (IATSs).

The data set consists of measurements from two total stations and images acquired by those two total stations to a drone/UAS. The UAS itself was equipped with a GNSS antenna, a laser scanner, an IMU, a nadir-looking camera. and four photogrammetric targets to allow the estimation of the UAS orientation based on IATS. The layout is shown in the included image Siegendorf_2026_Layout.jpg. It is complemented by extensive reference data in the study area. The reference data consists of TLS data and about 100 Ground Control Points that can be used for bundle-block adjustment of the camera data.

The specific sensors of this field study are:

• 2x Leica MS60 IATS
• RIEGL VUX-120²³ with integrated RiLOC-F and Sony ILX-LR1
• DJI M400 with adapted landing gear (Figure 2)
• Leica GS18 as GNSS base station

The data was acquired to study the full 6-DoF trajectory estimation of airborne platform with IATS. However, the data set can for example also be used for research towards mobile mapping, GNSS processing and decentralized multi-sensor systems using IATS. 

Technical details

• Image-assisted total station (IATS)
  •  IATS data consists of two types of data that were logged using a Raspberry Pi 4 controller and the network interface of the Leica MS60 IATS.
    • (i) polar observations (distance and angle measurements) with a sampling frequency of about 20 Hz. The data is provided raw, i.e. as extracted using the GeoCOM interface and processed (i.e. a latency of about 31 ms is corrected)
    • (ii) telescope images (see included image UAS_from_Telescope.jpg), the original sampling frequency was 30 Hz but we downsampled to about 5 Hz for the data publication to reduce the size of the iamge data. The images were acquired by logging an rtsp stream on the controller and timestamping the arrival time. The name of each image file is the timestamp in seconds of day. Our empirical evaluation found a latency of -150 ms w.r.t. the polar observation data. For a short flight segment, the horizontal and angular observations to the four photogrammetric targets on the UAS (visible in the IATS images) are also given
  • The setup coordinates, orientation unknowns and instrument heights of both IATS are given in setup.txt
•  GNSS
  • A Leica GS18 was set up as base station to record GNSS data.
• LiDAR
  • The LiDAR system was a RIEGL VUX-120

• IMU
  • The system uses a RiLOC-F IMU which is integrated with the LIDAR (Figure 3)

• Camera
  • The LiDAR system has a rigidly installed Sony ILX-LR1 to acquire nadir-looking images

• UAS
  • The DJI M400 was tuned by adding a 1 m long landing gear beneath the original landing gear. On the landing gear, four spherical targets (3D printed) were installed to allow the orientation estimation by IATS images
  • The UAS is shown in detail in the included image file UAS_Setup.jpg but without the prims installed.

• General
  • In this data set, the trajectory was flown four times, twice with 3 m/s and twice with 5 m/s speed.
  • 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

Data structure

The data set consists of Raw Data (e.g. RiNEX files, ASCII observations from the IATS, and images), Processed Data (Cartesian measurements from GNSS and IATS as well as extracted target observations by the IATS), and Reference Data (TLS, Ground Control Points, Coordinates of Fixed Points). The file crs_definitions provides the information to transform into global coordinate system as well as the definitions of the body frame to sensor frame transformation. The Raw Data containts *.rdbx files, the can be read using the RDBLib, but all basic information is also in the .las files.

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.