Visual and LiDAR-based for the mobile 3D mapping

Multi-sensors fusion to construct an accurate 3D map about unknown indoor scene and outdoor in the simultaneous localization and mapping (SLAM) area is becoming increasingly popular. In this paper our methods pay an attention to how to optimize the 2D map from the indoor or outdoor unknown large-scale environment and how to make the 3D map intuitive and accurate. Firstly, this paper accomplished the calibration between 2D Light Detection And Ranging sensor (LiDAR) and panoramic camera, using horizontal 2D LiDAR to finish 2D SLAM. Then the new method that closed-loop detection was based on visual was used to assist the 2D SLAM to get a better optimization result in sub-maps efficiently. Finally, vertical LiDAR data was added to get the 3D point cloud, using the images from a panoramic camera to accomplish the registration with point cloud. The result of final 3D map with RGB information demonstrated that our methods are feasible.