Virtual validation, precise simulations, and new regulatory requirements are driving the digital transformation of vehicle development. Validation and conformity with standards and regulations must be an integral part of the development process right from the start – through to homologation. The next logical step: the complete implementation of digital homologation.
Rapid Change Through Software
The automotive industry has been undergoing rapid change for some time now. Manufacturers and suppliers are increasingly focusing on innovative vehicle architectures such as software-defined vehicles (SDV), where software is the key success factor. At the same time, test requirements are increasing, particularly in the areas of ADAS and autonomous driving. Studies show that vehicles with autonomous driving systems have to complete ten to one hundred times more mileage than comparable vehicles without these functions in order to obtain approval from the authorities.
Furthermore, the integration of complex software systems makes general vehicle development more difficult, while at the same time, there is market pressure to reduce development cycles. In China, for example, OEMs report development times of 18 months for a completely new vehicle. To keep up with this pace, a reliable homologation strategy that starts early on is necessary. Digital homologation offers a solution for this: It accelerates validation and certification processes and, at the same time, meets increasing quality and safety requirements.
Digital Homologation vs. Virtual Homologation
Digital and, in particular, virtual homologation processes are becoming increasingly important in order to meet the growing requirements for safety, regulation, and speed. But how do they differ?
Digital homologation describes the entire process for obtaining type approval. As can be seen in Figure 1, in addition to all legal and functional requirements in the development process, it also takes into account the subsequent assessment by the technical service. This approach enables end-to-end traceability of vehicle properties and software functions, as well as efficient change management. Digital evidence serves as the basis for audits and approval.
In contrast, virtual homologation refers to the implementation of validation and verification steps in the V-cycle using virtual test environments. Precise simulation models can be used to simulate real scenarios in software-in-the-loop (SIL) or hardware-in-the-loop (HIL) simulators. In addition, as is also shown in Figure 1, regulatory requirements are taken into account, which are described in the so-called Simulation Credibility Framework. This framework, explained in more detail below, defines the conception, planning, integration, and evaluation of models and simulations (M&S). This means that software updates, parameter changes, and the resulting system behavior can be checked before over-the-air (OTA) updates are performed and forwarded to the technical service for re-approval. This reduces physical testing, saves costs, and speeds up the time to market.
Regulations with the Option of Virtual Homologation
With the implementing regulation EU R2022/1426, based on the basic regulation (EU) 2019/2144, a clear framework for virtual homologation was created for the first time. The Simulation Credibility Framework defined therein specifies the requirements that must be met for the models and simulation (M&S) to be considered credible and accepted by the technical service for the type approval of autonomous driving functions. The Simulation Credibility Framework, therefore, serves as a systematic approach to assess the trustworthiness of M&S and is described in detail in the regulations. At the heart of this requirement is the Simulation Handbook, describing the entire lifecycle of tool-chain use: from design and integration through calibration and validation to monitoring.
At the end of 2024, the "World Forum for Harmonization of Vehicle Regulations", WP.29 for short, included the virtual homologation approach in UNECE R171 (Driver Control Assistance System – DCAS), thus opening the door to virtual tests for Level 2 systems as well. In 2025, the same committee approved an amendment to UNECE R152 (Advanced Emergency Braking System – AEBS), intended to enable the use of virtual test environments for the homologation of emergency brake assistants. The official announcement of this step is expected in 2026. In addition, there are currently discussions about allowing virtual tests for all relevant regulations in the field of ADAS – provided that the defined process is adhered to.
The Virtual Homologation Process
With the Simulation Credibility Framework, the DCAS regulation (UNECE R171) defines the regulatory framework to assess the credibility of models and simulations (M&S). dSPACE has developed processes based on this framework, shown in Figure 2 and with which these M&S can be validated in accordance with the regulations.
First, the car manufacturer selects the relevant functions that are to be implemented in the software, system architecture, and vehicle behavior. These functions are then partially tested in virtual test environments such as HIL or SIL. The generated virtual measurement data is compared with real data and validated accordingly. The results are reproducibly documented.
A central component is the provision of the necessary documents: the system description, the test reports in accordance with ISO 26262 and ISO 21448 (SOTIF), a safety manual, and a description of the virtual test tool chain. These documents are essential for the traceability and verifiability of the results.
The technical service checks the virtual test tool chain and audits its correct use. Safety-relevant vehicle functions and the performed SIL and HIL tests are evaluated. A positive evaluation verifies the credibility and consistency of the virtual environment. This procedure ensures that the M&S used for homologation are not only created and executed in a technically correct manner, but can also be used for the type approval of vehicles for regulatory purposes.
Figure 2: Safeguarding the function and the virtual tool chain. DCAS: Driver Control Assistance System.
This makes the virtual homologation process not only a technical, but also a regulatory foundation for the approval of modern vehicle systems.
dSPACE Expertise for Reliable Virtual Homologation
Virtual homologation is a complex topic. In addition to detailed technical knowledge, it also requires a high level of regulatory understanding and a great deal of practical experience in validation. This is precisely where dSPACE Consulting's strengths lie. The consulting team supports customers with the question of how the Simulation Credibility Framework can be efficiently integrated into existing test concepts. The customers benefit from a customized test strategy to integrate this legal framework. dSPACE supports them in creating a well-founded simulation handbook, which serves as a guideline for the entire M&S development and utilization process.
This is complemented by the dSPACE team's expertise in designing proven model validation workflows that ensure simulations remain credible and traceable.
In addition, dSPACE already has a dedicated model simulation portfolio. The basic scenarios required by the DCAS regulation (UNECE R171) and the extended test scenarios have already been simulated virtually. The model portfolio, therefore, provides an immediately usable basis for the tests required by the regulatory authorities. But the real traffic scenarios do not end with the basic scenarios alone: A lane change, for example, can take place on different roads, for example, with or without structural separation, oncoming traffic, or overtaking vehicles. dSPACE fulfills this requirement for dynamic environmental conditions with its ISO 26262-certified simulation tool chain. With established tools such as ModelDesk, AURELION, and ASM, scenarios can be flexibly created, expanded, and realistically modeled.
As already mentioned, the credibility of the sensor models is essential for virtual homologation. Because only if the virtual vehicle in the simulation behaves exactly the same as a vehicle on a real test track can this serve as valid proof. AURELION creates sensor models that seamlessly combine physical reality with virtual testing and meet the requirements of UNECE R171. Among other things, dSPACE has developed a standardized workflow for the validation of vehicle models based on ISO 7401 and ISO 22140. As shown in Figure 3, measurement data from real driving tests is automatically compared with the results of the simulation, and the quality of the simulation is summarized in a report using comparison plots and KPIs. The figure shows an example of a direct comparison of the yaw rate and lateral acceleration, with the dark blue signal representing data from a real test drive, and the light blue signal representing data in the dSPACE tool "ASM Vehicle Dynamics".
With this combination of strategy, methodological expertise, and practical tools, dSPACE accompanies and supports its customers through the entire virtual homologation process. In this way, dSPACE creates the basis for bringing vehicle innovations to the road faster, safer, and in compliance with regulations.
About the Authors
Florian Sontheim
Florian Sontheim is a Senior Consultant at dSPACE in Paderborn, Germany
Ahmet Karaduman
Ahmet Karaduman is a Consultant at dSPACE in Munich, Germany
Advantages of virtual homologation
- Reduces time-to-market
- Meets regulatory requirements at an early stage and in an audit-proof manner
- Reduces homologation costs through virtual test procedures