‘ADApproved’ project completed: New test platform for automated driving

Technische Hochschule Ingolstadt (THI), together with its partners, has completed the “ADApproved – Autonomous Driving Sensor Homologation Lab” project. At its heart is a test facility capable of simulating spray water under controlled conditions. It is intended to help make automated driving functions more reliable and safer in the future.

From left to right: Professor Werner Huber, Letícia Cristofoli, Thomas Hempen, Bernhard Killermann, Jakob Reckenzaun and Nigel Bruce (THI, AVL and SETLabs; photo: THI).

Thick spray, poor visibility, and barely discernible obstacles pose major challenges for automated vehicles. To investigate such situations in a laboratory setting, the project partners developed an automated test environment for modern vehicle sensor technology.

From January 2023 to June 2026, THI, SETLabs Research GmbH (as consortium leader), AVL Software and Functions GmbH, and AUDI AG (as an associated partner) worked together. The closing event took place at AVL’s sensor weather laboratory in Roding.

THI designed and built a system capable of precisely generating spray conditions of varying intensity. In addition, a semantic LiDAR dataset was created, enabling a systematic analysis of sensor and object detection performance.

“Autonomous vehicles must function reliably not only in sunshine, but also in rain and in conditions of restricted visibility. With our test facility, we can specifically simulate such conditions and investigate them under identical conditions. This lays an important foundation for more reliable driver assistance and automation systems,” says Professor Werner Huber, Head of the Institute for CARISSMA(C-IAD).

The facility combines sensor technology, actuators, and specially developed control software. This allows test procedures to be automated, measurement data to be fully recorded, and test series to be repeated at any time under the same conditions.

The investigations show that the spray significantly impairs environmental detection. LiDAR sensors produce less meaningful point clouds, meaning that objects are sometimes detected late, incorrectly, or not at all. The results provide valuable insights for the further development of automated mobility.

During the closing event, the system was successfully demonstrated in AVL’s weather chamber. In the future, it is to be used in further research projects to test new sensor systems under challenging weather conditions.

Back