Regresar

On-board visual SLAM on a UGV using a RGB-D camera

Abstract:

We present a approach to real-time localization and mapping using a RGB-D camera, such as Microsoft Kinect, and a small and powerful computer Intel Stick Core M3 Processor. Our system can run the computation and sensing required for SLAM on-board the UGV, removing the dependence on unreliable wireless communication. We make use of visual odometry, loop closure and graph optimization to achieve this purpose. Our approach is able to perform accurate and efficient on-board SLAM, and we evaluate its performance thoroughly with varying environments and illumination conditions. The experiments demonstrate that our system can robustly deal with difficult data in indoor and outdoor scenarios.