LiDAR SLAM for indoor exploration using Unmanned Aerial Vehicles

The role of Unmanned Aerial Vehicles, UAVs, in the context of disaster response has traditionally been to extend the senses of the responders by providing a bird's eye view of the affected area. In contrast, the role of indoor exploration has often been the purview of Unmanned Ground Vehicles, UGVs, since their greater payload allowed them to use the heavier sensors needed for such tasks. The recent development of lighter, cheaper, and more accurate sensors has opened the possibility of deploying small UAVs to map indoor spaces. UAVs offer many advantages over UGVs; they are more agile and quicker, they can fly over obstacles, and reach structures inaccessible from the ground. The goal of this work has been to identify the best hardware and software setup to perform Simultaneous Localization And Mapping, SLAM, in indoor spaces using a small UAV. To this end we performed an extensive battery of tests in simulated environments using a custom plugin we developed for this purpose. Having selected the most promising solution we then implemented drift correction strategies to increase the robustness and accuracy of the solution.