구글 티처블 머신을 활용한 군사장애물 분류 모델 설계 및 실증 연구

• Published in: 한국CDE학회 논문집 Vol. 27; no. 2; pp. 137 - 147
• Main Authors: 황재민(Jaemin Hwang), 마정목(Jungmok Ma)
• Format: Journal Article
• Language: Korean
• Published: (사)한국CDE학회 01.06.2022; 한국CDE학회
• Subjects: 기계공학; Deep learning; Obstacle; Modeling and simulation; Teachable machine; Obstacle clearance tank
• ISSN: 2508-4003; 2508-402X
• DOI: 10.7315/CDE.2022.137

With the recent development of Obstacle Clearance Tank (K-600) that can overcome minefield, rockfall and road crator, ROK Army can shorten the time required to overcome obstacles and increase operation efficiency. However, in order to overcome the lack of military service resources in the future and be guaranteed to survive operator, Unmanned Obstacle Clearance Tank should be introduced along with artificial intelligence technologies. In order to develop the Unmanned Obstacle Clearance Tank, the initial recognition stage is critical among “recognition- control-action” stages. This study aims to build the obstacle recognition and classification model based on Google teachable machine and verify the model using the real RC-car camera test environment. KCI Citation Count: 2