Universal path planning for an indoor drone

During the construction and maintenance of building, universal path planning for an indoor drone navigation is needed. There are many two-dimensional (2D) path planning methods, but they are not appropriate for a three-dimensional (3D) indoor environment with many obstacles in it. In this study, we present a novel approach to plan universal paths for drones in a known indoor environment using a voxel model. This approach can make the drone fly at some distance from the obstacles by computing a 3D buffer around the obstacles, using our algorithm 3D propagating approximate Euclidean distance transformation (3D PAEDT). Two types of paths are presented using A* and distance transformation algorithms: safe shortest path (SSP) and safe least cost path (SLCP). Both paths ensure that the drone maintains a minimal distance from the obstacles. SLCP ensures that the drone flies at a fixed height. Several experiments are conducted to test the approach in a two-story building.