The evolution of today’s ADAS sensor stack is largely driven by new legislations, consumer ratings that are increasing out of distribution detection and robustness of the systems. More specifically, there is a need for increased performance in challenging driving conditions such as glare, darkness, or adverse weather. In this presentation, we will go through a test campaign with several scenarios performed in various challenging conditions using different sensing modalities. Results and comparisons will be provided in what could be named dimensioning use-cases. Additionally, an innovative approach to utilize the strengths and complement the challenges with different sensor types involving early fusion with Machine Learning models. The presentation also addresses how synthetic data can support and speeds up development in this domain. The objective was to address the needs for future sensing and system approach for increased automation.

Join this session to learn more about:

• Sensing modalities test campaigns with several scenarios performed and conditions
• How to use sensing modes involving early fusion with ML models
• How synthetic data can support the development of sensors as well as increased automation