Research on active SLAM with fusion of monocular vision and laser range data

This paper proposes an effective method for extracting corner features on structured environment based on sensor fusion of monocular vision and laser range data during Simultaneous Localization and Mapping (SLAM). Fusing vision and laser data of the same corner feature, not only improving the accuracy of SLAM, but also obtaining more three-dimensional information, which extends a two-dimensional map building by laser data to be three-dimensional. Feature matching based on images solve data association problem better than laser range data only. In addition, the accuracy of SLAM can be improved by using active exploring strategy. Simulation and experimental results show the effectiveness of the proposed method.