SAFE AI ENGINEERING
AI engineering ensures safety of AI components for autonomous, connected driving
Start: 08/2021
End: 12/2024
All SofDCar Alliance project partners are focusing their research on the challenges of future E/E & SW architecture in vehicles. The vehicle is understood as part of a networked vehicle and system environment. Its integration will be enabled by means of a “Data Loop” and based on a new type of “Digital Twin” to allow for digital sustainability (of existing and future vehicle generations) as well as efficient data use and innovative use cases over the entire life cycle of the vehicle (re-deployment).
The FZI is contributing its extensive research expertise in the field of mobility. The focus of SofDCar is on validating, verifying, and maintaining the consistency of vehicle variants, recognizing limitations and weaknesses in the driving function, evolving both variants and individual functionalities, and managing the resulting versions with the help of the digital twin. The spotlight is also on quality-assured application processes across vehicle boundaries. In addition, applied IT security issues such as the security assurance of vehicle components will be studied using Artificial Intelligence methods, including practical security tests on selected components. In the Test Area Autonomous Driving Baden-Württemberg (TAF BW), various aspects of software-over-the-air updates, validation, and increased robustness are being tested.
This project is part of a government-sponsored research program funded by the Federal Ministry for Economic Affairs and Climate Action (BMWK).
Intelligente Lösungen für den Transport von Menschen und Gütern für die Mobilität der Zukunft stellen einen Schwerpunkt der FZI-Forschung dar. Dazu entwickelt das FZI integrierte Mobilitätssysteme – von der Fahrzeugautomatisierung über die Anwendung von KI in Verkehrssystemen bis zur urbanen Mobilität und Logistik.
Funding notice:
The SofDCar project is funded by the Federal Ministry for Economic Affairs and Climate Action (BMWK).
Project partners:
At the Karlsruhe Institute of Technology (KIT) and the FZI Research Center for Information Technology, we research concepts for developing and securing software-implemented functionality in vehicles and their cloud infrastructure. The focus is on security analyses, dealing with a high number of versions and variants, and over-the-air updates.
Together, we are examining various aspects of software-over-the-air, safeguarding, and robustness enhancement at the Autonomous Driving Test Field Baden-Württemberg. Our test vehicles are also used for this purpose. Our goal is to continuously improve the safety of AI-based functionalities, also with the valuable evaluation opportunities weaned from the test field.
AI engineering ensures safety of AI components for autonomous, connected driving
Data trustee model protects use of individual health data in research
AI-based software detects and corrects motion artifacts in OCT images.
Thanks to the measurement of social interaction, psychological stress caused by loneliness can be detected at an earlier stage
Digital platform for dementia prevention helps address the lack of exercise
More exercise in everyday life: Self-learning real-time system recognizes behavior patterns and recommends customized activities
AI assistant serves as first point of contact for legal questions
AI engineering empowering security reasoning throughout the entire lifecycle of an AI function
Secure and self-determined digital identities in e-commerce
Co-design for application-specific microelectronics with innovative chiplets in Baden-Württemberg