The University of Freiburg and Bosch Research are partnering on the AI-Drive project, focused on developing the next generation of AI algorithms for automated driving, according to Prof. Dr. Abhinav Valada, project leader and member of the Faculty of Engineering at the University of Freiburg. The three-year project, which will include ten doctoral students, is aiming to create a more unified framework where different modules are closely connected and mutually optimized, representing a significant departure from traditional approaches to technological advancements in automated driving.
- Bosch Research is providing €3.7 million in funding, as well as real test vehicles, and the project is also intended to help strengthen research on automated driving in Germany.
- Research will focus on achieving a closer coupling of prediction and planning modules in order to mimic human driving behavior, avoiding overly cautious or aggressive driving maneuvers.
- The project will also aim to develop innovative methods for neural architecture search to automate the design and optimization of network architectures, thus improving the efficiency of the development and deployment cycle.
- AI-Drive will be following a “white box” approach, characterized by transparency and the ability of humans to interpret intermediate results, as this is expected to foster trust and simplify certification processes.
- The scientific community is also expected to benefit from the project, as technological and theoretical innovations resulting from this collaboration will be published in leading scientific journals and conferences.
Prof. Dr. Claudius Gläser from Bosch Research stated that the partnership with the University of Freiburg is an important link to cutting-edge research in automated driving and will enable advancement in the state of research, in addition to integrating it into future automation solutions. Valada, along with Prof. Dr. Frank Hutter and Prof. Dr. Joschka Bödecker from the Faculty of Engineering and the ELLIS Unit Freiburg, will also be involved in the project. Overall, the goal of these developments is to lead to a safer, more transparent, and more robust overall autonomous driving system.