Image processing for a more accurate GNSS-based positioning in urban environment - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Image processing for a more accurate GNSS-based positioning in urban environment

Résumé

Most of the Intelligent Transport Systems rely on GNSS positioning information. Unfortunately, most of the vehicles circulate in dense urban areas where the GNSS signals are affected by multipath caused by the buildings surrounding the antenna. The CAPLOC project deals with inaccuracy by associating image processing techniques and signal propagation knowledge to understand and reduce GNSS inaccuracy. Our solution embeds fisheye cameras located on the roof of a vehicle and oriented upwards to capture images of the sky. Our approach consists in processing the images to 1/ determine whether satellites are located in a sky or non-sky region and 2/ construct a 3D model around the camera/antenna in order to estimate the pseudo-range delays caused by the close propagation effects. This knowledge allows us to propose several strategies of exclusion/weighting or correction of these pseudoranges for a better accuracy.
Fichier non déposé

Dates et versions

hal-01471581 , version 1 (20-02-2017)

Identifiants

  • HAL Id : hal-01471581 , version 1

Citer

Juliette Marais, Sébastien Ambellouis, Cyril Meurie, Julien Moreau, Amaury Flancquart, et al.. Image processing for a more accurate GNSS-based positioning in urban environment. 22nd ITS World Congress, Oct 2015, Bordeaux, France. 12p. ⟨hal-01471581⟩
119 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More