Skip to Main content Skip to Navigation

Véhicules Intelligents : Etude et développement d'un capteur intelligent de vision pour l'attelage virtuel

Abstract : If many aspects of our life became more pleasant thanks to the use of leading-edge technologies, a long time was necessary for the transport field to fill its delay on the matter. Today, this progress became a necessity. This work describes the environment perception in front of the vehicle, using a stereoscopic sensor based on the concept of smart sensors, whose aim consists carrying out a Virtual Attachment between two vehicles.
After a presentation of the associated problematic, the first chapter draws up the state of the art as regards intelligent vehicles. The second one introduces the concept of intelligent sensor and presents the designing approaches; we apply to identify the various services and functionalities that this intelligent stereoscopic sensor must integrate to contribute to the accomplishment of the Virtual Attachment task.
The last chapter describes the realization of the stereoscopic sensor. We detail the problems raised when applying stereovision to the field of transport and the solutions we propose. Thus, we discuss difficulties of calibration procedure, real time extraction of region of interest and the problem of certification of the obtained data. We have implemented an extraction and tracking feature to satisfy temporal requirements. Performances of each module of our sensor are supported by experimental results obtained in actual situations. Finally, we present a technique allowing us to follow a vehicle just by using a single camera.
Document type :
Complete list of metadata

Cited literature [111 references]  Display  Hide  Download
Contributor : Mohamed Zayed <>
Submitted on : Thursday, April 12, 2007 - 9:54:07 AM
Last modification on : Tuesday, November 24, 2020 - 2:18:22 PM
Long-term archiving on: : Friday, September 21, 2012 - 2:06:08 PM


  • HAL Id : tel-00141213, version 1



Mohamed Zayed. Véhicules Intelligents : Etude et développement d'un capteur intelligent de vision pour l'attelage virtuel. domain_other. Université des Sciences et Technologie de Lille - Lille I, 2005. Français. ⟨tel-00141213⟩



Record views


Files downloads