Social Infrastructure Deterioration Diagnosis AI System Actively Using Ground Penetrating Radar Images Without Teacher Labels

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K11879
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61010:Perceptual information processing-related
• Research Institution: Oita National College of Technology
• Principal Investigator: kimoto Tomoyuki 大分工業高等専門学校, 電気電子工学科, 教授 (30259973)
• Co-Investigator(Kenkyū-buntansha): 園田 潤 仙台高等専門学校, 総合工学科, 教授 (30290696)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 地中レーダ / 劣化診断 / ディープラーニング / 教師無し学習

Outline of Final Research Achievements

Transportation infrastructure such as roads and bridges, which were built during the construction boom during the period of high economic growth, has deteriorated over the past 50 years, leading to accidents such as cave-ins and collapses. Although it is necessary to prevent these from occurring, since there are a huge number of roads and bridges all over the country, it is necessary to efficiently find problems such as internal cavities. In this research, by emitting radio waves into the ground and using AI to identify the reflected waves from the ground, we developed a system that can accurately identify risk factors such as cavities that occur inside the ground.

Free Research Field

AI応用

Academic Significance and Societal Importance of the Research Achievements

道路などの社会インフラの大規模劣化が進んでいるわが国では、低コスト、高速かつ高精度で識別し、ピンポイントで修繕が必要な部分を探し出す必要がある。そこで、本研究では、電磁波や音波を用いた非破壊検査で、社会インフラの劣化診断を行う手法の高精度化を試みた。その結果、従来法より識別精度を向上させることができることが分かってきた。これにより修正に必要な税金を抑えることができたり、危険個所を早期に特定して未然に事故が発生するのを防いだりする方向に前進したことになる。