The level 3 driver monitoring system is primarily designed for autonomous driving operations, with the aim of determining the driver's capability to take over the driving task and enabling a smooth driving task handover interaction between the autonomous system and the driver. The system employs image processing and AI deep learning algorithms to continuously monitor three physiological signs, including head movement, gaze direction, and eye closure, within a 30-second timeframe, in order to meet the driver availability requirements of UN R157. In the event that the system detects that the driver has lost the ability to take control or the vehicle requires the driver to take control, the system will issue a takeover request for the driver to take over or enter a minimal risk operation.

2. Alert issued if the driver leaves the position for more than one second. 

3. Head pose detection range: yaw +70° to -70°, pitch +30° to -30°. 

4. Meets the requirements for driver availability recognition under UN R157. 

5. Camera resolution is 720P.

6. Can operate on an AI edge computing platform with a processing speed ≥ 10FPS.

This technology can be applied in various types of vehicles equipped with in-cabin monitoring system, such as car, bus, intercity bus, autonomous vehicle…etc.

This technology can make domestic industry who want to develop driving monitoring system can get abilities about AI face detection, gaze tracking, driving availability recognition. Acquiring these techniques can enhance product competitiveness, and expand the international market.