This paper describes the new Lane Detection module which is now operative on the ARGO autonomous vehicle and enables the vehicle to drive itself on roads and highways. It is only based on the processing of a monocular sequence of images acquired from the moving vehicle. A first simpler version, tested and demonstrated in a 2000+ km tour throughout Italy in 1998, showed some problems which have now been eliminated by the current approach. The paper describes how the new algorithm can adapt to different road and environmental conditions, as well as how it can reconstruct scenes which are partly occluded by other vehicles or in which lane markings are partly missing.
Titolo: | Visual Perception and Learning in Road Environments |
Autori: | |
Data di pubblicazione: | 2000 |
Abstract: | This paper describes the new Lane Detection module which is now operative on the ARGO autonomous vehicle and enables the vehicle to drive itself on roads and highways. It is only based on the processing of a monocular sequence of images acquired from the moving vehicle. A first simpler version, tested and demonstrated in a 2000+ km tour throughout Italy in 1998, showed some problems which have now been eliminated by the current approach. The paper describes how the new algorithm can adapt to different road and environmental conditions, as well as how it can reconstruct scenes which are partly occluded by other vehicles or in which lane markings are partly missing. |
Handle: | http://hdl.handle.net/11381/1449482 |
ISBN: | 1586030787 9781586030780 |
Appare nelle tipologie: | 4.1b Atto convegno Volume |
File in questo prodotto:
File | Descrizione | Tipologia | Licenza | |
---|---|---|---|---|
ias-6.pdf | Documento in Pre-print | PUBBLICO - Pubblico senza Copyright | Administrator Richiedi una copia |