The future of automotive development
The automotive industry is facing a major challenge
Driver monitoring cameras must not only function reliably to ensure safety systems such as drowsiness detection and distraction warnings – they are now also legally mandated. At the same time, semi-automated systems from Level 2++ upwards rely heavily on the precise attention classification provided by these cameras. Meeting these requirements is essential for vehicle type approval and thus for the safety of all road users.
The high proportion of traditional vehicle tests continues to lead to rising costs, long test phases, and reduced flexibility. In an era of ever-shorter development cycles and increasing cost pressures, new approaches are needed to make testing and validation even more efficient and to avoid recursive processes. This is exactly where we come in: Our solution allows for the virtual testing of driver monitoring systems – quickly, cost-effectively, and scalably. As a development partner, we support our customers in complying with legal requirements and driving innovation – agile, smart, and trusted.
Virtual tests instead of hardware marathons
IR-based camera systems must reliably detect gaze direction, eyelid movements, and other physiognomic features even under changing conditions – from bright sunlight to nighttime lighting. The key lies in synthetic datasets that enable comprehensive virtual human subject studies.
We are developing a high-performance and scalable test platform that generates video data to simulate realistic near-infrared camera recordings. This allows for the virtual testing of recognition states such as attention, distraction, and microsleep – without the need for actual vehicles. Dynamic scenarios with virtual test subjects are created, which, thanks to our expertise, can be quickly developed, flexibly modified, and precisely adapted to test requirements. This enables the testing of thousands of variations even in early development phases: different age groups, diverse appearances, types of glasses and beards, varying sitting and head positions, and different pupil sizes.
This approach makes securing the software component not only more efficient but also significantly more cost-effective. Furthermore, design studies can be implemented flexibly, quickly, and without physical setup. Our EVOs then interpret, classify, and evaluate the results. With their extensive experience in interior sensor technology, we achieve the ideal balance between innovation, regulatory compliance, and reliability.
Key innovation:
• RGB-to-NIR transformation for realistic video data
• Realistic scene generation with any digital subjects
• Automated protection at the touch of a button – overnight
This allows critical perception states such as fatigue or distraction to be tested virtually under changing light conditions while driving – without real vehicles.
Curious? We'd be happy to show you how virtual testing can accelerate your development processes and reduce costs. Contact us!
Visualization, based on a modified 3D modelOriginal model: n.brizitskaya via Sketchfab. License: Creative Commons Attribution (CC BY)No connection to Porsche AG.
