Microdrone-Based Indoor Mapping with Graph SLAM

Unmanned aerial vehicles offer a safe and fast approach to the production of three-dimensional spatial data on surrounding space. In this article, we present low-cost SLAM-based drone for creating exploration maps building interiors. The focus is emergency response mapping in inaccessible or potentially dangerous places. For purpose, used quadcopter microdrone equipped with six laser rangefinders (1D scanners) an optical sensor positioning. employed SLAM designed map indoor spaces planar structures through graph optimization. It performs loop-closure detection correction recognize previously visited places, correct accumulated drift over time. proposed methodology was validated several environments. We investigated performance our against multilayer LiDAR-carrying macrodrone, vision-aided navigation helmet, ground truth obtained terrestrial scanner. experimental results indicate that system capable quality small spaces, handling problem. without loop closure average 1.1% (0.35 m) 31-m-long acquisition trajectory. Moreover, comparison demonstrated flying provided comparable LiDAR-based given low deviation between point clouds built by both drones. Approximately 85 % cloud-to-cloud distances were less than 10 cm.