Indoor mobile robot navigation using doorplate landmarks. (29th January 2007)

• Record Type: Journal Article
• Authors: Zou, An-Min; Hou, Zeng-Guang; Tan, Min; Gupta, Madan M.; Nikiforuk, Peter; Homma, Noriyasu
• Is Part Of: International journal of vehicle autonomous systems. Volume 4:Number 2/3/4(2006)
• Page Start: 143
• Page End: 155
• Publication Date: 2007-01-29
• Subjects: camera calibration -- colour image segmentation -- localisation -- mobile robots -- neural networks -- vehicle autonomous systems -- autonomous vehicles -- robot navigation -- doorplate landmarks -- indoor navigation -- landmark recognition -- robot vision -- visual sensors -- floor landmarks; Autonomous vehicles -- Periodicals; Motor vehicles -- Technological innovations -- Periodicals; Motor vehicles -- Electronic equipment -- Periodicals; Robots -- Control systems -- Periodicals; Intelligent transportation systems -- Periodicals; 629.20285
• Languages: English
• ISSNs: 1471-0226

Abstract

In the autonomous mobile robot industry, the landmark-based navigation method has been widely used in which landmark recognition plays an important role. The landmark recognition using visual sensors relies heavily on the quality of image segmentation. To localise and navigate the robot using vision information, the camera has to be first calibrated. In this paper, we use the camera to recognise floor and doorplate landmarks. Neural networks are used to calibrate the camera and solve the colour image segmentation problems, respectively, then the doorplate number recognition algorithm is described and the topological map is created by treating the doorplate landmarks as topological nodes.