»AI platform concept for storage and processing of learning and test data«
Autonomous driving has the potential to revolutionize road traffic, but at the same time it presents society with enormous challenges.
Potential of autonomous driving
On the gradual path to autonomous vehicles, machine processes are making increasingly complex decisions. They take over tasks that were previously reserved for humans. These decisions are learned by means of artificial intelligence (AI) based on observations. A major advantage is that technical systems can ensure a significantly higher and more constant level of attention, which leads to a significantly higher freedom from error and thus to greater safety for road users as a whole.
Learning data as a basis for AI development
To realize the potential of AI for autonomous driving, the "learning data set" is of crucial importance. For training and successful learning of the various AI subfunctions, it must have sufficient scenario coverage. Only in this way can AI-based driving functions and services function safely, robustly and reliably in practice.
Data platform for AI learning data
The project aims to design a pre-competitive database to which all OEMs and suppliers operating in Germany can contribute annotated data from real drives and simulations according to a uniform scheme and which is available to industry, scientific partners and administration alike. The Institute for Legal Informatics is significantly involved in the research of the legal framework for a corresponding platform.
Funded by the federal ministry of education and research.
Project manager |
Professorial chair for Civil Law, Legal Informatics, German and International Business Law and Legal Theory – Prof. Dr. Georg Borges |
Project representative |
Research associate Andreas Sesing (This email address is being protected from spambots. You need JavaScript enabled to view it.) |
Project partner |
Bundesanstalt für Straßenwesen (BASt) |
Project duration |
September 2018 – March 2019 |