“Simulation Credibility as a Critical Enabler for Automated Driving”
- Natalie Julia Weiss
- Jun 10
- 4 min read
Updated: Oct 9
Source: ATZ - Automobiltechnische Zeitung
Car manufacturers are increasingly leveraging simulation technologies to advance the development of automated driving systems. Thomas Guntschnig from MORAI explains why a standardized verification of simulation credibility is so important in this context.
The development costs for automated driving are enormous. The automotive industry is therefore relying on simulation technologies to accelerate development processes. However, the reliability of simulation results is a challenge for the industry. How can the credibility of simulation-based validation be ensured?
Simulation plays a central role in the development and validation of automated driving systems. “To ensure simulation outputs are reliable, standardized and transparent evaluation criteria are essential. This is exactly what we are currently working on together with the IPEK institute at KIT and with TÜV Süd as an independent authority: an assessment framework (“Credibility Assessment Framework”) that defines when a simulation is considered sufficiently trustworthy. It is crucial that the scenarios are modeled realistically, that the sensor technology is correctly virtualized, and that the results are reproducible. Only then can simulation truly serve as a valid basis for the safety assessment of automated vehicles and subsequently also be used in the ADS homologation process.
You already mentioned that South Korean simulation specialist MORAI and the Institute for Product Development (IPEK) at the Karlsruhe Institute of Technology (KIT) are developing a Credibility Assessment Framework that aims to standardize the credibility of complete simulation toolchains for low- to highly automated vehicles (SAE levels 2 to 4+). TÜV Süd has been appointed to independently review the results. What is the goal of the project?
The project aims to establish a clear benchmark for classifying and evaluating the trustworthiness of simulation toolchains – from scenario creation and component modeling to realistic simulation execution. The aim is to develop a framework that is both technically sound and recognized by regulators, and that is generic enough to be applied to various simulation software solutions and ADS systems. For this reason, we have involved TÜV Süd as a neutral authority to independently review the results of the evaluation methodology. In that way, we want to ensure that simulation gets officially applied as part of the safety assessment of ADS systems in the homologation and type approval process in the future.
What role does the credibility of simulations play when it comes to regulations compliance and safety assurance?
Simulation is becoming increasingly important for the validation of highly automated driving functions. This is due to the growing complexity of these systems, which have to guarantee a high level of safety in challenging traffic situations. Simulation makes it possible to test a wide variety of scenarios – including those that, for safety reasons, are not feasible to be tested in real-world environments.
The credibility of the simulation is a key factor here. Only if the virtual test mirrors the real driving situations as closely as possible, the simulations are representative and can therefore guarantee the safety of ADS systems. A uniform assessment framework provides the necessary clarity and makes the use of simulation not only more efficient but also legally secure.
How do Korea and Germany differ strategically in their approaches and strategies regarding self-driving cars?
In Korea, technological innovation is often driven in a pragmatic way and implemented in pilot projects. New systems are tested quickly and developed iteratively. In Europe, and also in Germany, the approach is more standards- and safety-driven. This takes more time to transfer systems into regular operation, but delivers very robust results. Both approaches are valid and complement each other well: While Korea often moves faster regarding the implementation, Germany brings enormous depth in engineering and in system validation.
How does the German and European automotive industry compare to other global players in the field of automated driving?
Mercedes-Benz was the first company in the world to receive approval for an SAE Level 3 system with its Drive Pilot. When it comes to highly automated SAE Level 4 vehicle solutions, the US and China – led by companies such as Waymo and WeRide – are gaining ground, and Europe has to intensify its efforts to keep pace.
Germany maintains a strong position in the areas of systems engineering, standards development, and functional safety. At the same time, we see that Asian countries are often quicker to put new technologies into practice. Germany has extensive technical expertise and the highest quality standards, but this often means that processes take longer. Simulation can help to pick up the pace here without compromising on quality or safety.
When will autonomous driving become part of our everyday life?
Autonomous driving will not become mainstream overnight, but targeted use cases are steadily emerging – in logistics, for example, with automated trucks on highways or in passenger transport using robotic shuttles in cities. In my opinion, such use cases will become increasingly visible over the next 5 to 10 years. Europe is well positioned in terms of regulation: The UNECE is currently pushing ahead a global harmonization, which will be completed by 2026 and adopted by national policies. However, infrastructure expansion and social acceptance are also essential for its introduction. Simulation can accelerate this change – in development, validation, and as a tool for testing new mobility concepts in a societal context.
Article Source:
Interviewer & Author: Christiane Köllner
Media Outlet: ATZ - Automobiltechnische Zeitung / Springer Professional
