RCS Profiles
Testing of automated driving functions becomes more and more virtualized. A key part of this development is the implementation and validation of perception sensor models. With this dataset, radar cross-sections (RCS) of multiple objects with different aspect angles are examined. The experiment is set up as a slalom course to measure the RCS under realistic conditions in a real world environment.
The dataset including all meta data is available for free from TUdatalib.
Test Setup
The tests where set up as a slalom course on the August-Euler-Airfield in Darmstadt, Germany. The slalom was set up in a reproducible manner by placing markers and pylons on the test track. The ego vehicle and the objects drove through the slalom at a constant speed. Each test drive contains 10 slalom periods and for each object 10 test drives were conducted.
The green pylons are gates through which the object passes and the orange cones are markers over which the object drives. The grey doted line represents the object's trajectory. The S-Class drives through the green gates in a straight line to measure different sensor azimuth and object yaw angles.
Sensors
The ego vehicle was a Mercedes S-Class equipped with the following perception and reference sensors:
- 6 Continental ARS408 Radars
- 1 Velodyne VLP32C Lidar
- 1 Blickfeld Cube 1 Lidar
- 1 Ibeo LUX 2010 Lidar
- 1 Genesys ADMA G-Pro+ RTK-GNSS & IMU in each vehicle
Objects included in Dataset
BMW 535
BMW i3
BMW Z3
Honda Accord
Mercedes Unimog
Opel Astra
Opel Corsa
Toyota Auris
VW Beatle
VW Caddy
VW Crafter
VW Golf
VW T5
Analysis Code
To analyze the dataset, several Matlab tools are available. Check out our open source repository.
Citation
If you find this data set useful for your research, please consider citing the publication:
L. Elster, M. F. Holder and M. Rapp, "A Dataset for Radar Scattering Characteristics of Vehicles Under Real-World Driving Conditions: Major Findings for Sensor Simulation," in IEEE Sensors Journal, vol. 23, no. 5, pp. 4873-4882, 1 March1, 2023, doi: 10.1109/JSEN.2023.3238015
FZD Contributors
Lukas Elster
Research Associate
Martin Holder
Former Research Associate
Manuel Rapp
Student Assistant