i4Driving: Integrated 4D driving modelling with consideration of uncertainties

The project is funded by EU funds from the Horizon Europe Research & Innovation Framework Programme and supported by the European partnership Connected Cooperative and Automated Mobility (CCAM).

Connected Cooperative and Automated Mobility (CCAM), also known can be as “networked, cooperative and automated mobility”, is considered one of the next major trends in the automotive industry. This approach focuses on creating a more user-centred and integrative mobility system aimed at improving road safety, reducing traffic congestion and minimizing the ecological footprint.

In the i4Driving project, Aschaffenburg UAS is collaborating with 13 partners from seven European countries. The project is supported by the CCAM partnership and brings together all the expertise needed to achieve its goals, such as traffic engineering, human factors, data and computer science. Additionally, the project team works with a strong international network of academic and professional partners to collaborate and coordinate new approaches.

The goal of the i4Driving project is to develop an industry-oriented methodology that provides a reliable and realistic foundation for the virtual evaluation of CCAM systems in road safety. The evaluation benchmark is the comparison of CCAM systems' behavior with that of safely and efficiently controlled vehicles, where driving behavior is modeled after experienced and knowledgeable drivers in the reference models.

To compare the safety performance of autonomous vehicles with human-driven vehicles, the full range of human driver performance will be captured through critical driving simulations.

The following concepts will be applied:

•  A multi-level, modular and extensible simulation software that integrates both existing and new models of human driving behaviour,

in combination with

•  An innovative, cross-disciplinary methodology to account for the significant uncertainty in both human behaviour and various use cases.

The uncertainty will be assessed to check how well the model inputs, parameters and structure are justified. It will also help evaluate how experts from different fields interpret the results and how valid the underlying assumptions are. The project team has experimental tools to realistically simulate (near) accidents in scenarios involving multiple drivers (with access to data sources, advanced driving simulators and field laboratories).

Further information is available on the i4Driving project website.