The project RELAI - Risk Estimation with a Learning AI is funded by the Federal Ministry of Transport and Digital Infrastructure (BMVI). With our cooperation partners IPG Automotive GmbH, Fraunhofer IOSB and the University of Stuttgart, the EDI hive Framework is used to generate a catalog of synthetic test scenarios for the development of autonomous vehicles and to make them available to the public via a web portal. Here, AI algorithms evaluate real, challenging traffic situations in mixed traffic and convert them into synthetic scenarios. The main focus is on the behavior of pedestrians and cyclists in critical situations as well as the expectations of these road users on the behavior of autonomous vehicles. Another important goal is to be able to automatically calibrate different simulation environments (including a virtual reality (VR) environment) through the synthetic test scenarios. It also indicates in which areas or sections specific test scenarios can be carried out in real road tests. The automatically generated test scenarios are made accessible to the public via a web portal on the EDI hive platform. In addition, the EDI hive framework is directly linked to the mCloud, the data portal of the BMVI, so that this project also helps to build up a comprehensive mobility database in Germany.
See also at Electric mobility south-west.