Lane marking aided vehicle localization

Author

Tao, Zhenning ; Bonnifait, Philippe ; Fremont, Vincent ; Ibanez-Guzman, Javier

Author_Institution

Univ. de Technol. de Compiegne (UTC), Compiegne, France

fYear

2013

fDate

6-9 Oct. 2013

Firstpage

1509

Lastpage

1515

Abstract

A localization system that exploits L1-GPS estimates, vehicle data, and features from a video camera as well as lane markings embedded in digital navigation maps is presented. A sensitivity analysis of the detected lane markings is proposed in order to quantify both the lateral and longitudinal errors caused by 2D-world hypothesis violation. From this, a camera observation model for vehicle localization is proposed. The paper presents also a method to build a map of the lane markings in a first stage. The solver is based on dynamical Kalman filtering with a two-stage map-matching process which is described in details. This is a software-based solution using existing automotive components. Experimental results in urban conditions demonstrate an significant increase in the positioning quality.

Keywords

Global Positioning System; Kalman filters; computer vision; image matching; sensitivity analysis; 2D world hypothesis violation; LI-GPS estimates; automotive components; camera observation model; digital navigation maps; dynamical Kalman filtering; lane marking aided vehicle localization system; lane markings detection; map matching process; sensitivity analysis; video camera; Accuracy; Cameras; Global Positioning System; Roads; Shape; Vehicles;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Transportation Systems - (ITSC), 2013 16th International IEEE Conference on

Conference_Location

The Hague

Type

conf

DOI

10.1109/ITSC.2013.6728444

Filename

6728444