Artificial intelligence-based driving control system uses cameras to determine if the driver is ready to take over.
It is true that in automated driving the car is the boss, but in most cases, it still depends on the driver. Most modern high-tech vehicles are already equipped with driver monitoring systems, but they hardly use any camera image data and are mainly limited to fatigue detection.
The Fraunhofer Institute for Optronics, System Technologies, and Image Exploitation IOSB created the world’s first system that can use image data to infer driver behavior and evaluate the speed at which they control a car. The system is intended to interact between the driver and the vehicle and is used in conjunction with a camera located in the cabin.
The system uses artificial intelligence-based algorithms and analyzes camera data in real-time to see if the driver is making a call, playing with a child, or checking a mobile phone.
The driving control system extracts images of the driver or passenger to form a digital skeleton-a kind of stick figure that replicates the posture of a person’s body. Therefore, it uses skeletal motion and complementary object recognition to infer activity. This helps to analyze how distracted the person is and how long it takes to refocus on the road and remind the driver to control the car.
The AI driving control system is compatible with regular visible-light cameras, infrared night-vision cameras, or depth-sensing 3D cameras.
The technology respects the privacy and therefore complies with strict EU regulations and a high level of data protection awareness. The recorded video will not be transmitted outside the car, and no personal data will be collected.