Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Verkehrssituationsabhängige fahrzeugautarke Fahrerunterstützung - Zur Integration von Maschinensehen und Fahrer-Fahrzeug-Kommunikation

Situation adapted autonomous driver support - towards integration of computer vision and driver-vehicle-communication
 

VDI-Gesellschaft Fahrzeugtechnik:
Elektronik im Kraftfahrzeug. Tagung 1994 = Electronic systems for vehicles
Düsseldorf: VDI-Verlag, 1994 (VDI-Berichte 1152)
ISBN: 3-18-091152-2
pp.511-521 : Abb.,Lit.
VDI-Gesellschaft Fahrzeugtechnik (Tagung) <1994, Baden-Baden>
German
Conference Paper
Fraunhofer IITB ( IOSB) ()
Autonome Fahrzeugführung; autonomous road vehicle guidance; driver-vehicle interaction; Fahrer-Fahrzeug-Wechselwirkung; road traffic situation; Sichtsystem; systembasierte Fahrerunterstützung; Verkehr; vision-based driver assistance; vision system

Abstract
Within various projects at the IITB, two road vehicles have been equipped and commissioned with systems for automatic lateral and longitudinal guidance which are based on computer vision: a small truck (Mercedes Benz 609D) and a sedan (BMW 735iL). The navigation system Travelpilot has been integrated into the MB 609D in cooperation with Robert Bosch GmbH. This enables a driver support system to link two kinds of on-board knowledge: the static knowledge about the road network for vehicle location and navigation, and the dynamic knowledge about the current traffic situation next to the vehicle. For that purpose, hints about the current traffic situation are gathered by realtime image sequence analysis. Automatic rides on obstacle free roads with speeds up to the maximal speed of the truck (more than 90 km/h) and at least up to the recommended maximal speed of the German Autobahn (130 km/h) with the BMW 735iL verified our approaches as well as our implementations. The consolidated methods have recently been ported to a smaller, but more powerful vision system. Thus the preconditions are met to start systematic experiments in driver-vehicleinteraction without the necessity to average over a large variety of traffic situations. On the contrary it is now possible to use automatic image sequence analysis methods in combination with information from the navigation system in order to relate the driver's actions and reactions to the current traffic situation. We hope that the experiences gathered in these investigations will lead to some kind of automatic driver assistance that is adapted to the driver and the situation, and thus increases the applicability and the acceptability of such systems.

: http://publica.fraunhofer.de/documents/PX-39736.html