According to MEMS Consulting, researchers at the Fraunhofer Institute in Germany are developing a camera and radar combination module that responds 160 times faster than human drivers. This project is called KameRad and aims to bring higher safety to autonomous driving.

The Fraunhofer Institute for Reliability and Microintegration (Fraunhofer IZM) is collaborating with industry partners (InnoSenT, Silicon Radar, Jabil Optics Germany, AVL, John Deere) and research institutions (Fraunhofer Institute for Open Communication Systems FOKUS, DCAITI) to develop this camera-radar combination module to capture changes in road traffic conditions faster.

This combination module is about the size of a smartphone, with a response time of less than 10 milliseconds. According to a study by the University of Michigan, this response time makes it 50 times faster than existing sensor systems and 160 times faster than an average human driver.

The true innovation of the new system lies in its powerful integrated signal processing capability, allowing all signal processing to be done directly within the module. The system can selectively filter data from radar and stereo cameras, enabling immediate signal processing or purposefully delaying it for subsequent processing stages.

Irrelevant information can be identified but will not be forwarded. Through sensor fusion, data from cameras and radar are integrated. Then, using neural network technology, the data is evaluated based on machine learning to determine the actual traffic impact. As a result, the system no longer needs to send state information to the vehicle but only needs to respond to commands, freeing up the vehicle bus to process important signals, such as detecting children suddenly running onto the road.

“Integrated signal processing significantly shortens response time,” explains Christian Tschoban, head of the RF and Smart Sensor Systems department at the Fraunhofer Institute.

Tschoban and his colleagues are working together on the development of the KameRad project. The functional demonstration module they developed features stereo cameras with two cameras on the left and right, resembling a “gray box.” The project will continue until 2020, when project partners AVL List and DCAITI will test the initial prototypes of the module, including road tests in Berlin.

Tschoban hopes that in a few years, this “gray box” will become standard equipment in every vehicle, providing more safety for the automation of urban traffic.

Recommended Training:

The “3D Imaging and Sensor Training Course” will be held from July 19 to July 21 in Wuxi. This course invites excellent instructors from the 3D vision industry chain to analyze 3D imaging and sensing technology based on core components: (1) Infrared light source technology, in-depth analysis of VCSEL; (2) MEMS beam manipulation technology covering various MEMS mirror principles (electrostatic, electromagnetic, thermal, piezoelectric); (3) Liquid lenses; (4) Diffractive optical elements (DOE); (5) High-performance photodetectors: SPAD/SiPM; (6) Photodetector arrays: ToF image sensors; (7) Optical MEMS and sensor simulation technology.If you are interested, please contact:

MEMS Consulting

Contact: Zhao Tingting

Phone: 18021192087

E-mail: [email protected]