The SET‑Level project ran from 3 March 2019 to 31 October 2022 and was financed by the German Federal Ministry of Economics and Energy under grant number 19A19004P. It followed the PEGASUS initiative and brought together roughly 20 partner firms, including OEMs, suppliers, technology developers and research institutions. The consortium’s core objective was to advance simulation‑based development and testing for automated driving functions, with a particular emphasis on standardising test‑description formats, simulation interfaces and data exchange. ETAS GmbH, a subsidiary of Bosch with more than 2 000 employees, led several work packages and contributed its expertise in mathematical solvers, co‑simulation middleware and ADAS/HAD toolchains.
Technically, the project produced a suite of artefacts that are now available to all partners. In work package 2.3 the team defined the requirements for a simulation environment and led five sub‑packages: solver, performance, SIL, HIL and connected systems. The outcome was a set of documents, checklists and recommendations that codify minimum solver specifications, performance guidelines and interface requirements for SIL, HIL and networked systems. These artefacts were derived from extensive workshops, individual and joint activities, and survey results, ensuring that they reflect the needs of the entire consortium.
Work package 3.1 focused on modelling. After completing the methods, processes and templates in 3.1.1, the team moved to 3.1.2 where it implemented two sensor models—wheel‑speed and inertia—and set up two environment models using the ETAS COSYM co‑simulation middleware combined with CARLA (open‑source) and aiSim (aiMotive). The simplified weather model that had been planned for the second half of 2021 was discontinued following discussions with partners. The modular approach, with open interfaces, allows the same middleware to be used with different simulation tools, thereby reducing integration effort and improving reproducibility.
In work package 4.4 the consortium instantiated toolchains for development and testing, demonstrating how the standardised artefacts can be integrated into real‑world workflows. The project also evaluated its results against practical use cases and demonstrators, confirming that the new standards and tools improve test coverage and reduce cycle times. The cloud‑based simulation services developed as part of the project enable rapid deployment of standard tests, addressing the challenge that excessive modularisation and a proliferation of tools can otherwise slow down testing.
The funding enabled a comprehensive, technically deep collaboration that would have been difficult for a single company to achieve alone. By aligning the interests of OEMs, suppliers and research bodies, the project produced high‑quality, standardised data formats and tools that are now ready for industrial deployment. The insights gained—particularly in solver requirements, performance optimisation and the balance between modularity and system speed—will inform ETAS’s product strategy and provide a foundation for future advances in autonomous vehicle verification and validation.