SPACE CONSTRUCTION METHOD OF MINE RESCUE ROBOT BASED ON IMPROVED ORB-SLAM2

After a sudden safety accident in a mine occurs, the mine rescue robot needs to collect visual data in the mine environment and perform functions such as spatial construction of the emergency area. Using the ORB-SLAM2 algorithm of RGB and RGB-D, the scalability optimization research was carried out. OCTOMAP was selected to optimize the map, and the mine rescue robot was used to experiment in the real indoor scene to verify the effect of the method in this paper. The results show that the space construction effect obtained is more applicable than the real-time point cloud construction, reduces the processing memory space, provides a method for the space construction of the mine rescue mobile robot, and has important practical significance for the subsequent research on the mobile positioning of the mine robot.