Development of a Tightly Coupled Vision/GNSS System

Bernhard M. Aumayer, Mark G. Petovello, Gérard Lachapelle

Abstract:	This paper focuses on the loose and tight integration of a stereo-vision system with a Global Navigation Satellite System (GNSS) based navigation system. A simplified vehicular dynamic model was chosen and implemented in order to gain maximum advantage from the vision sensor’s information. Data collected in open-sky as well as urban environments was processed in post-mission with the University of Calgary’s GSNRx™ software GNSS receiver. The results include performance measures of the loose and tight integration approaches using high-quality GPS-synchronized computer vision cameras, which are compared to the output of a GNSS-only solution as well as a tactical-grade reference trajectory. More than 20,000 images and 30 minutes of raw GNSS measurements in a dynamic scenario with a total travelled distance of more than 10 km were collected, processed and analyzed, and the resulting errors in the trajectories were compared.
Published in:	Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014) September 8 - 12, 2014 Tampa Convention Center Tampa, Florida
Pages:	2202 - 2211
Cite this article:	Aumayer, Bernhard M., Petovello, Mark G., Lachapelle, Gérard, "Development of a Tightly Coupled Vision/GNSS System," Proceedings of the 27th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2014), Tampa, Florida, September 2014, pp. 2202-2211.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In