인공지능 기반 전시 활주로의 불발탄 및 폭파구 정량화에 관한 연구

Abstract

At the beginning of the war, the runway of a military airfield becomes the top target for enemy missile strikes, and crater of various sizes are formed on the attacked runway, while many unexploded ordnance bombs remain. According to the existing airfield damage analysis technology to solve this problem, manual survey of unexploded ordnance and crater by reconnaissance team relying on visual observation and measuring tape measure, transmission of information about the size and location of crater by radio to the facility control room, and detonation by the facility control room Classification of old buildings, manual selection of MOS, and runway repair (damage restoration) process by the damage recovery team are sequential is done. However, the existing technology takes a long time to work due to the manual surveying of unexploded ordnance munitions and crater, and there is a high possibility that the reference point will be lost. There is a problem, and there is a possibility of secondary work by the information morning moon in the runway repair process. A fixed FOD(Foreign Object Debris) detection system can be applied to solve this visual inspection problem, but there is a problem in that the installation of a fixed tower itself is not preferred around the runway due to flight safety issues. This study is proposed to solve the above problems, and based on drone footage, artificial intelligence algorithm, and SfM(Structure from Motion) algorithm for point cloud generation, it is possible to analyze the damage at the emergency airfield in wartime and to develop an automated system for repairing damage to the flight pavement. Its purpose is to develop.