Authors: Liwei Han, De Xu, Yi Zhang
Addresses: China North Optical-electrical Technology Co., Ltd., Beijing, 100176, China. ' The Key Laboratory of Complex Systems and Intelligence Science, Institute of Automation, Chinese Academy of Sciences, Beijing, 100190, China. ' College of Automation, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China
Abstract: When a mobile robot navigates in an indoor environment using visual dead reckoning method, its positioning accuracy suffers from accumulated errors. Therefore, it is necessary to use landmarks to make correction. This paper investigates the natural landmark-based localisation for an indoor mobile robot. The landmarks used here include smoke detection sensors, speakers and lights on the ceiling that are widely available in many offices and corridors. To improve the real-time performance, the proposed method utilises global and local strategies to search lines on the ceiling, as well as the line fitting algorithm based on Hough transform and random sample consensus. The pose of mobile robot is estimated with visual dead reckoning method, and then corrected via PnP-based positioning method with natural landmarks. Experimental results verify the effectiveness of the proposed methods.
Keywords: landmark recognition; random sample consensus; RANSAC; visual positioning; dead reckoning; mobile robots; robot localisation; robot positioning; indoor robots; natural features; ceiling features; self-localisation; robot navigation; positioning accuracy.
International Journal of Modelling, Identification and Control, 2010 Vol.10 No.3/4, pp.272 - 280
Published online: 10 Aug 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article