Problem statement

Autonomous systems in traffic must be able to react safely to a wide range of possible situations in open, dynamic environments, including rare and unpredictable events. Existing methods are reaching their limits, particularly in complex, unpredictable operating areas. On the road, situations such as pedestrians suddenly appearing in blind spots, unexpected construction sites, faulty or contradictory traffic signals, extreme weather conditions affecting sensors, or technical malfunctions in the vehicle are at the forefront. These scenarios are currently hardly systematically covered. In the rail sector, there are additional challenges, such as people or obstacles on the tracks.

Objective

The diversity and dynamics of open environments overwhelm traditional safety methods. CONTROL therefore takes a new approach: the system continuously assesses how safely it can operate in a given situation and responds to uncertainties with cautious measures, such as adjusting speed or controlled evasive maneuvers. CONTROL builds on the PEGASUS and VVM projects, which systematically derived test cases and secured complex operating areas. As part of the VDA's flagship initiative for autonomous and connected driving, CONTROL extends this path with the ability to actively assess and manage uncertainties. At FTM, we are developing methods for high-precision localization based on multimodal sensor fusion as part of the project. We analyze and compensate for uncertainties arising from environmental influences and near-series sensor technology to ensure robust and precise vehicle localization.

Funding authority