Improved quantitative driver behavior models and safety assessment methods for ADAS and AD (QUADRIS)

Important results from the project

The automotive industry needs fast and accurate simulation-based safety assessment methods to a) be able to bring AD to market and b) stay competitive when developing improved ADAS. QUADRIS successfully adressed three knowledge gaps in simulation-based safety assessment methodology: a) how to improve on creation and validation of virtual safety assessment scenarios, b) a lack of driver response models for run-of-road scenarios and c) the lack of driver reference models for AD development.

Expected long term effects

The project has filled critical knowledge gaps in how to credibly conduct simulations aimed at assessing the safety benefit of AD and ADAS systems that are under development. These methodological improvements are expected to directly benefit the automotive industry, as they can be applied directly within ongoing development projects.

Approach and implementation

The project was mainly carried out in the form of three doctoral theses. These have, unusually, both followed the schedule and the intended direction. Some obstacles regarding data access and handling of analysis results have needed to be sorted out along the way, but no impossible obstacles have arisen. The collaboration has worked excellently, just as in previous collaborative projects between the parties involved.