METHOD FOR THREAT SITUATION AWARENESS USING ONTOLOGY AND DEEP LEARNING IN UAV

摘要

The present invention relates to a method for recognizing a threat situation of an unmanned aerial vehicle, comprising the steps of: collecting information on at least one first object existing around the UAV; generating a first grid map based on the information on the first object; step, analyzing the first grid map using an ontology-based first inference model, and first inferring a relationship between the UAV and the first entity, and using a second inference model based on deep learning and verifying a result of the primary inference, and secondary inferring a relationship between the UAV and the first entity. Accordingly, the present invention can more accurately recognize the threat situation of the unmanned aerial vehicle.