Road Spray in Radar and Lidar Data
Simulation-based testing supports the challenging task of safety validation of automated driving functions. Virtual testing always entails the modeling of automotive perception sensors and their environment. In the real world, these sensors are not only exposed to global weather conditions, like rain, fog, snow etc., but environmental influences also appear locally. Road spray is one of the more challenging occurrences, because it involves other moving objects in the scenario. This data set is designed to systematically analyze the influence of road spray on lidar and radar sensors. It consists of sensor measurements of two vehicle classes driving over asphalt with three water levels to differentiate multiple influence factors.
The dataset including further meta data is available for free from TUdatalib.
One lane is watered by a distribution system on a Unimog.
The water film dissipates over time enabling three water film levels: Fully covered, partially covered, moist.
Reference targets for lidar and radar are placed on the other side of the lane
There are two different kinds of logical scenarios:
1) The ego vehicle is positioned stationary next to the watered lane. The object vehicles are driving past the ego with different velocities and different water film heights.
2) The ego vehicle is following the object vehicles in different distances with different velocities.
The ego vehicle is a Mercedes S-Class equipped with the following perception and reference sensors:
- 1 Velodyne VLP32C Lidar
- 1 Ibeo LUX 2010 Lidar
- 1 Aptiv ESR 2.5 Radar
- 1 Genesys ADMA G-Pro+ RTK-GNSS & IMU in each vehicle
- 9 different object velocities (50 - 130 km/h)
- 3 water levels (Fully covered, partially covered, moist)
- 2 object vehicles (VW Golf, VW Crafter)
- 2 distances for dynamic scenarios
- Every individual scenario is recorded 3 times
If you find this data set useful for your research, please consider citing the publication:
(Currently under review at IEEE)
C. Linnhoff, D. Scheuble, M. Bijelic, L. Elster, P. Rosenberger, W. Ritter, D. Dai, H. Winner: "Simulating Road Spray Effects in Automotive Lidar Sensor Models" in IEEE Sensors, 2022
Former Head of FZD