With a complete package of sensor systems, semi-autonomous or in the future, autonomous vehicles should reliably register their environment. In addition to cameras, radar and ultrasonic sensors, lidar sensors will be used to offer vehicles the best all-round view. In the project “iLIDS4SAM”, eleven Austrian partners gather relevant knowledge under the leadership of Infineon Austria.
Current driver assistance systems are still designed for simple traffic scenarios
Given Stefan Rohringer, head of the development center at Infineon, it is “a showcase project that encompasses the entire technology, application and test chain”. The consortium has a research volume of almost 5.7 million euros, co-financed by the Federal Ministry of Climate Protection, Environment, Energy, Mobility, Innovation and Technology (BMK) as part of the “ICT of the Future” program. Project partners include AVL List GmbH, ams AG, Silicon Austria Labs GmbH, Graz University of Technology and Graz Competence Center Virtual Vehicle.
Radar and camera systems are already replacing the human eye in a number of vehicles. Lidar system is another key technology for fully automatic driving. In Graz, Infineon is developing microelectronic solutions for these laser-based “sensory devices”, which are to be installed behind the windscreen, in the headlights or in the taillights to control the vehicle’s surroundings in combination with radar or camera systems.
High demands on complex city traffic
However, the current driver assistance systems are still designed for relatively simple traffic scenarios such as motorway traffic or parking assistance without pedestrians, cyclists or cross-traffic. Predictable hazard detection in complex urban traffic, however, places much higher demands.
The iLIDS4SAM (Integrated LiDAR Sensors for Safe & Smart Automated Mobility) project aims to meet this challenge. Lidar is an abbreviation for “Light Detection and Ranging” and works in a similar way to radar: Instead of radio waves, laser beams are emitted in the infrared range to detect objects in the long range of the vehicle using light and to determine their distance. A laser beam uses microchip mirrors to scan the surroundings with millimeter precision, horizontally and vertically. Distance, position and speed for road users and objects are calculated with the time it takes for the light to reach the object and back. And here the Austrian project consortium has recognized a lot of development potential.
The goal is to develop a powerful and inexpensive laser sensor system with “deep learning” data management. This is intended to make the vehicle an “intelligent and expectant road user”, as it was called in the Infineon broadcast. To make the sensors compact, robust and fully integrated, the corresponding components, connection technologies and also the mirror design should be optimized and further miniaturized. In test runs, the researchers want to collect a large amount of real data in order to be able to predict behavior using signal processing algorithms and derive a risk assessment. The compact sensor system will be tested in urban and rail traffic as well as in agricultural applications.
A data management plan connected to the open source platform of the European Research Center CERN has already been implemented. Based on this publicly available data pool, further research projects, development and improvements should be able to begin.
“This is about a significant further development of key technologies for safe autonomous driving. We bring together excellent Austrian partners from science and industry and thereby strengthen European technical expertise in global competition,” concluded Sabine Herlitschka, CEO of Infineon Technologies Austria AG.
(WHAT / red, Photo: WHAT / WHAT (dpa))