Multi-Modal Detection of Driver Impairment: Exploring Radar-Camera Fusion for Safer Roads

Purpose and goal

The aim is to evaluate camera-radar fusion for real-time detection of impaired drivers. The project will demonstrate the potential of the method and lay the foundation for a full-scale development project that could lead to a robust and reliable system for DUI detection in vehicles.

Expected effects and result

The project is expected to provide important technical and societal impacts through the validation of radar for DUI detection and the development of a multimodal radar-camera solution with high precision and reliability. This can contribute to reducing the risk of traffic accidents and related fatalities while strengthening the competitiveness of Swedish companies in advanced vehicle surveillance.

Planned approach and implementation

Technical, regulatory and operational eligibility requirements as well as business conditions for the project are defined. This is followed by technical validation of radar-based DUI detection, which forms the basis for the subsequent phase with multimodal sensor fusion where radar and camera data are combined to improve detection capabilities. Finally, the system´s potential is evaluated, obstacles are identified and a roadmap is developed for full-scale implementation and further development.