Lane recognition self-learning scheme of mobile robot based on integrated perception system

Author

Yang Yi ; Zhu Hao ; Fu Meng-Yin ; Wang Mei-ling

Author_Institution

Sch. of Autom., Beijing Inst. of Technol., Beijing, China

fYear

2013

fDate

23-26 June 2013

Firstpage

1046

Lastpage

1051

Abstract

In this paper, a kind of integrated perception system for mobile robot is presented, which consists of 3D Lidar, 2D camera and their spatial registration. Based on the system and support vector machine (SVM), a self-supervised learning scheme between 3D point cloud data and 2D image data has been established, which can identify the traversable lane in driving environments through data association and parameters training. With this approach, vision-based autonomous navigation can be achieved and its effectiveness has been verified by extensive robot experiments.

Keywords

cameras; image fusion; image registration; mobile robots; navigation; optical radar; support vector machines; unsupervised learning; visual perception; 2D camera; 2D image data; 3D Lidar; 3D point cloud data; SVM; data association; driving environments; integrated perception system; lane recognition self-learning scheme; mobile robot; parameter training; robot experiments; self-supervised learning scheme; spatial registration; support vector machine; traversable lane identification; vision-based autonomous navigation; Cameras; Laser radar; Mobile robots; Navigation; Support vector machines; Three-dimensional displays;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Vehicles Symposium (IV), 2013 IEEE

Conference_Location

Gold Coast, QLD

ISSN

1931-0587

Print_ISBN

978-1-4673-2754-1

Type

conf

DOI

10.1109/IVS.2013.6629604

Filename

6629604