Development of a Packaging-Technology for a Camera–Radar-System for Autonomous Driving

About this Session
The achievement of fully autonomous driving requires further technological steps in the development of sensors, artificial intelligence and smart infrastructure.  On sensor level, systems must be reliable and robust in sensing to achieve high safety standards. Camera sensors fail in foggy or dusty environment. For radar sensors the separate detection of small targets which are very close to each other, such as pedestrians, is one of the mayor challenges. Especially for urban environments, this requirements lead to radar sensors with a very high angular resolution (approx. less than ±1.5°) in near and medium range (up to 100 meters). 

In a common research project partner from industry and research developed a new sensor concept based on a combined camera-radar-system. The challenges in the project are as follows: 

1)    Development and integration of a high resolution radar system 

2)    Development and integration of a high resolution camera system

3)    Integration in one package

For the high resolution radar-system a novel glass-substrate based scalable MIMO – Radar fronted with integrated antenna was developed (challenge 1). The angle resolution can be fixed in one dimension by choosing the number of frontends. The radar sensor is placed in the center of the sensor system, framed by two high resolution stereo cameras. Therefore the total width of the module is given by the minimum interspace between cameras. The cameras are connected through a Serializer directly with the artificial intelligence unit (challenge 2). 

Both sensors as well as the AI processing unit are integrated with a very high density (challenge 3). Sensors and electronics are integrated in metal housing for protection reasons. 

In this presentation an overview about the development steps of the package development and results of the final measurements of the system will be presented.

Contact Speaker

Download Slides