Semantic segmentation based mapping systems for the safe and precise landing of flying vehicles

Unmanned Aerial Systems (UAS) are a promising technology for many areas, including transportation, agriculture, inspection, and rescue missions. However, to enable a high level of autonomy, including Beyond Visual Line of Sight (BVLOS) filght, the drones should be able to perform safe landings in unknown areas without an operator. Hence there is a need for development of safe landing methods for autonomous drones. The autonomous UAVs can often be operated more economically than the conventional manned aircraft. As technology advances, autonomous UAVs are expected to play an increasingly important role in a variety of industries and applications. In this paper we have explored a semantic segmentation-based approach for the problem of autonomous landing.