A combined camera-radar module that reacts 160 times faster than humans
The highly integrated electronic system developed in the KameRad joint project combines the advantages of optical monitoring with radar measurement technology in a single system. The camera radar module developed by Fraunhofer IZM registers changes in road traffic significantly faster and has a reaction time of less than ten milliseconds. This means it reacts around 50 times faster than conventional sensor systems and 160 times more quickly than a human being.
Integrated signal processing shortens response time enormously
The data from the radar system and the stereo camera are processed and filtered directly in or on the module. Non-relevant information is recognized but not passed on[TC1] . Sensor fusion combines the data from the camera and radar. Based on neural networks, this data and, thus, various traffic conditions are evaluated using machine learning. As a result, the system does not send any status information but only reaction instructions to the vehicle. This leaves the vehicle's bus line free for important signals, such as a child suddenly running into the road.
Avoiding accidents in urban traffic
When a child runs into the road, it takes an average of 1.6 seconds for a person to hit the brakes. Autonomous vehicles equipped with radar and a camera system have a reaction time of 0.5 seconds. At a speed of 50 km/h, however, the car continues to drive unbraked for seven meters. The camera radar module, which is the size of a cell phone and has a reaction time of less than ten milliseconds, ensures that the car only travels 15 cm without braking - then the system reacts and sends signals to brake. This could prevent many accidents in urban traffic.
Find more information directly on the Fraunhofer IZM website: