Laser-based road recognition for a smart electric wheelchair

Author

Suzuki, T. ; Hashimoto, M. ; Takahashi, K.

Author_Institution

Grad. Sch., Doshisha Univ., Kyoto, Japan

fYear

2011

Firstpage

993

Lastpage

997

Abstract

This paper describes a method of road recognition, which is useful in outdoor environments, using a multi-layered laser range finder (MLRF) for a smart electric wheelchair. Road recognition means detecting the road in front of a wheelchair on which it can travel. The MLRF captures laser scanned images from the road surface and maps them onto a grid map. Paved roads are almost flat and have boundaries such as curbs, ditches or raised edges. We detect road flatness based on random sample consensus (RANSAC) method and estimate the shape of the road boundary based on a cubic (third-ordered) spline. Experimental results in our university campus show the feasibility of our method.

Keywords

edge detection; handicapped aids; laser ranging; random processes; roads; splines (mathematics); wheelchairs; MLRF; RANSAC method; cubic spline; grid map; laser-based road recognition; multilayered laser range finder; paved road; random sample consensus; road boundary; road flatness; smart electric wheelchair; Image edge detection; Mobile robots; Roads; Robot sensing systems; Shape; Wheelchairs;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics and Biomimetics (ROBIO), 2011 IEEE International Conference on

Conference_Location

Karon Beach, Phuket

Print_ISBN

978-1-4577-2136-6

Type

conf

DOI

10.1109/ROBIO.2011.6181417

Filename

6181417