Automated vehicles should be able to reliably sense traffic signs. Earlier systems, however, have had issues in understanding complicated traffic management with varied information about the course of the lanes or speed, as largely occurring on construction sites.
A team of researchers at Fraunhofer are building technologies for the real-time interpretation of such signs, which they will showcase at the CeBIT held in Hanover from March 20 to 24, 2017 (Hall 6, Booth B36).
Construction sites are a challenge for automated vehicles. This is due to the driving lanes becoming narrow, traffic jams developing, drivers frequently reacting under stress or insecurely, and accidents occurring more frequently.
The systems of the automated vehicles are also not capable of handling the complex situation: Old and new road markings overlap, and traffic cones and limiting beacons are hard to detect by the sensors. The signs contain varied information about the course of the lanes or permitted speed.
Recognizing Patterns More Quickly and Efficiently Using Deep Learning
"Our technology enables a system to read signs of this kind with a high degree of accuracy," says Stefan Eickeler, who is in charge for the subject of object recognition at the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS in Sankt Augustin, Germany.
The information is processed semantically, understood in terms of content, and made accessible for additional processing.
With Deep Learning – a key technology for the future of the automotive industry – we teach the software to recognize the classic patterns more quickly and efficiently.
Stefan Eickeler, Fraunhofer
Through the interplay between the navigation equipment and on-board computers, it may soon be possible for the distances to other vehicles to be maintained optimally, for differently designated highway exits on construction sites to be accurately identified, and for the speed to be modified in a timely manner.
What in the short term could be able to promote relaxation and increased safety when driving by means of assisted driving is intended to work all by itself in the long term: Automated vehicles will then react independently.
Stefan Eickeler, Fraunhofer
The Future Vision: Camera Replaces Numerous Sensors
An automotive camera is used which at present provides 20 to 25 frames per second. Simultaneously during the trip, these pictures are examined and information regarding signs, LED traffic signs, or lane information is identified and processed. Going forward, this camera will be able to operate as a primary interface, making several sensors redundant.
At the CeBIT, the Fraunhofer IAIS will utilize a virtual tour to exhibit numerous projects in the area of big data and machine learning, including the topics "Automated driving on construction sites", "Knowledge graphs for data-driven business models", or "Digital assistants and real-time recommendation systems"