A visual SLAM algorithm based on a novel artificial landmark system

Abstract

This paper presents a monocular vision and odometer based SLAM algorithm, making use of a novel artificial landmark which is called MR (Mobile Robot) code. A brief introduction of MR code system is given. A practical error model for odometric position estimation is described and the parameters are determined through experiments which verify the rationalities of the model. During robot motion, the information from visual observations is fused with that from the odometer by EKF. An actual experiment is carried out in indoor environment, the experimental results show the effectiveness of the proposed algorithm, which has improved the localization precision of the robot and the map accuracy.