In the AutoDevSafeOps research project, 14 research partners have been working since 1 October 2022 and will continue until 30 September 2025 to design building blocks for modular updates of safety-critical driving functions in connected vehicles for the software-defined vehicle and to test them in demonstrators with vehicle manufacturers. The aim of the project is to develop an integrated and adaptive DevOps framework that is specifically tailored to safety-critical automotive systems. DevOps is about making collaboration between software development, IT operations and vehicle fleets more efficient through shared processes and software tools.
Innovative approaches in AutoDevSafeOps
In AutoDevSafeOps, researchers and developers at asvin labs are working on combining traditional DevOps processes with new graph and AI methods for cyber risk analysis. Modern vehicles are software-defined vehicles consisting of several hundred million lines of software code. This complex software platform in the vehicle requires maintenance in the form of updates and patches throughout its entire life cycle. The AutoDevSafeOps project is researching and developing new models, tools and processes to ensure vehicle safety, cyber security and functionality in complex software systems. It considers software throughout the entire vehicle life cycle, from the development process and type approval to operation on the road.
AutoDevSafeOps
RESEARCH PARTNERS
Automotive manufacturers, among others, can benefit from the research results, for example through optimised processes in the approval procedure for new vehicle models at the Federal Motor Transport Authority (KBA). asvin labs is making important contributions here in research and development for automated test procedures in the field of cyber security for vehicles.
In AutoDevSafeOps, asvin is also focusing on the development of a comprehensive system for the verification and traceability of software artefacts throughout the entire software supply chain in the automotive industry. The resulting framework is intended to improve transparency, security and traceability from development to delivery via over-the-air updates. This will ensure complete insight into the origin, integrity and history of software components in a complex vehicle platform.
Results and applications
The results developed to date by asvin labs in the AutoDevSafeOps project can already support the approval process by providing reliable data points for integrity checks of the software installed on the vehicle and enabling detailed forensic analysis. These can also be used to verify compliance with norms and safety standards.
By taking into account the increasing complexity and modularity of modern automotive software, our contribution with asvin labs plays a crucial role in building trust and accountability in dynamic, safety-critical environments of modern vehicle architectures. In addition to the AutoDevSafeOps project, the partnership between asvin and DEKRA underscores the high relevance of automated auditing procedures in the field of homologation of new vehicle models in Germany and Europe.
The most important results of the AutoDevSafeOps project to date at a glance:
By embedding these functions in the DevOps pipeline, the results of this project not only improve confidence in software quality and thus the acceptance of SDV (Software Defined Vehicles) or connected cars. They also reduce the time and cost of certification and troubleshooting in the digital supply chain. This will contribute to a significant increase in development efficiency and regulatory security at KBA, OEMs and the automotive supply industry.
With the research and development work of asvin Labs, we support the safe and flexible delivery of innovations in connected cars and in Level 3 and Level 4 vehicles in semi-autonomous and autonomous driving.
Funding
AutoDevSafeOps is supported by funding from the German Federal Ministry of Education and Research.
