Share Email Print
cover

Proceedings Paper

Obstacle detection in range-image sequence for outdoor navigation
Author(s): M. Garduno; Bertrand Vachon
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

We deal with the conception of a perception system that's goal is to assist a mobile robot teleoperator by providing pertinent information about eventual obstacles appearing in the robot work space. This range image based perception system is to be embedded on a vehicle able to move at speeds up to 40 km/h in an outdoor environment. A method taking speed constraints into account is proposed. In the first step of this method, a segmentation algorithm is applied to the first range image scanned by the motionless robot to determine areas of interest. From these areas, distinctive attributes are computed and recorded as symbolic representations of each obstacle region. In the second and following steps, obstacles are localized in images scanned during robot motion. The difference between actual object position in the range image and its predicted value is used by an extended Kalman filter to correct the estimated robot configuration. A dynamic image segmentation using emergency and security criteria is carried out and new obstacles can now be detected from the range image and expressed in the robot coordinates system.

Paper Details

Date Published: 17 August 1994
PDF: 7 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182890
Show Author Affiliations
M. Garduno, Univ. of Compiegne (France)
Bertrand Vachon, Univ. of Compiegne (France)


Published in SPIE Proceedings Vol. 2357:
ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision
Heinrich Ebner; Christian Heipke; Konrad Eder, Editor(s)

© SPIE. Terms of Use
Back to Top