Innovation in Flux: Networked Thinking about Mobility

ACT4Transformation

ACT4Transformation
Securing AI-based systems in automated driving with a focus on data integrity and trustworthy algorithms.
Credit:

AI-generated with ChatGPT

As part of the ACT4Transformation project (Automated and Connected Technologies for Mobility Transformation), several institutes of the German Aerospace Center (DLR) are collaborating on key technologies for automated and connected driving. The DLR Institute of Transportation Systems is coordinating the project and developing concepts for integrating vehicles, infrastructure and remote operation. Meanwhile, the DLR Institute for AI Safety and Security is responsible for securing AI-based systems in automated driving, focusing on data integrity and trustworthy algorithms. Other contributions come from the DLR Institute of Systems Engineering for Future Mobility, which is working on digitally representing virtual bus stops, and Technology Marketing, which is concerned with the strategic transfer to service models. The 'Schwarzer Berg' neighbourhood in Braunschweig will serve as a real-life testing environment.

The project got underway in January 2025 and will continue for three years until the end of 2027. The first technological implementations in test operation are scheduled for 2026.

The project aims to ensure that automated driving is safe and reliable, even in complex traffic situations. This involves the flexible distribution of functions between the vehicle, infrastructure, and remote control centre; new concepts, such as virtual stops and crossings; and measures to enhance AI safety. The result is safer, more inclusive and more climate-friendly mobility for society. Politically, the project contributes to the implementation of the autonomous driving law and the German government's mobility strategy.
The project's recent launch, its high innovation potential, and the planned practical implementation make it a particularly relevant topic. DLR is therefore positioning itself strategically as a key player in future mobility technologies.

Contribution of the Institute for AI Safety and Security

Within the ACT4Transformation project, the DLR Institute of AI Safety and Security is leading central research efforts into the cybersecurity of automated transport systems. The project focuses on developing a cyber security management system tailored to the unique challenges posed by networked and data-driven mobility. The aim is to reliably protect data communication between vehicles, infrastructure, and control centres from manipulation, failure, and unauthorised access, thereby fostering trust in automated mobility.
One area of focus is securing AI-based components that will make safety-critical decisions in future transport systems. To this end, new attack scenarios, such as adversarial attacks, are analysed systematically, and countermeasures are developed in the form of detection, monitoring, and robustification strategies. Additionally, existing IT security standards are evaluated, and recommendations for their implementation in networked transport systems are provided.
In collaboration with other DLR institutes, we are developing practical protection concepts to enable the secure introduction of automated and connected vehicles in road traffic, covering everything from data flow to decision-making architecture. The effectiveness of the defence strategies developed is tested using simulations of realistic attack scenarios.
This initiative is a vital step in ensuring the mobility of the future is secure, resilient and trustworthy.

Participating DLR institutes and facilities

Contact

Dr. Michael Karl

Head of Department
German Aerospace Center (DLR)
Institute for AI Safety and Security
Safety-Critical Data Infrastructures
Rathausallee 12, 53757 Sankt Augustin
Germany

Karoline Bischof

Consultant Public Relations
German Aerospace Center (DLR)
Institute for AI Safety and Security
Business Development and Strategy
Rathausallee 12, 53757 Sankt Augustin
Germany