DocumentCode :
663983
Title :
Finding planes in LiDAR point clouds for real-time registration
Author :
Grant, W. Shane ; Voorhies, Randolph C. ; Itti, Laurent
Author_Institution :
Dept. of Comput. Sci., Univ. of Southern California, Los Angeles, CA, USA
fYear :
2013
fDate :
3-7 Nov. 2013
Firstpage :
4347
Lastpage :
4354
Abstract :
We present a robust plane finding algorithm that when combined with plane-based frame-to-frame registration gives accurate real-time pose estimation. Our plane extraction is capable of handling large and sparse datasets such as those generated from spinning multi-laser sensors such as the Velodyne HDL-32E LiDAR. We test our algorithm on frame-to-frame registration in a closed-loop indoor path comprising 827 successive 3D laser scans (over 57 million points), using no additional information (e.g., odometry, IMU). Our algorithm outperforms, in both accuracy and time, three state-of-the-art methods, based on iterative closest point (ICP), plane-based randomized Hough transform, and planar region growing.
Keywords :
Hough transforms; closed loop systems; distance measurement; image registration; iterative methods; optical radar; pose estimation; real-time systems; 3D laser scans; Hough transform; ICP; LiDAR point clouds; Velodyne HDL-32E LiDAR; closed-loop indoor path; iterative closest point; odometry; plane-based frame-to-frame registration; real-time pose estimation; real-time registration; robust plane finding; sparse datasets; spinning multilaser sensors; Equations; Iterative closest point algorithm; Laser radar; Laser theory; Measurement by laser beam; Sensors; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ International Conference on
Conference_Location :
Tokyo
ISSN :
2153-0858
Type :
conf
DOI :
10.1109/IROS.2013.6696980
Filename :
6696980
Link To Document :
بازگشت