Publications / 2020 Proceedings of the 37th ISARC, Kitakyushu, Japan

MLIT's Initiatives for Promotion the Efficient Construction and Inspection by using new Technologies such as AI and Robots in Japan.

Kenichi Watanabe
Pages 1356-1360 (2020 Proceedings of the 37th ISARC, Kitakyushu, Japan, ISBN 978-952-94-3634-7, ISSN 2413-5844)
Abstract:

The Ministry of Land, Infrastructure, Transport and Tourism (MLIT) is promoting the use of drones and other robots to improve the construction and management processes and improve the efficiency of infrastructure inspections, and is preparing an AI development environment that supports "human judgment." Specifically, MLIT is preparing "teaching data" which necessary for AI development and provides it to AI developers. In order to create an annotation rule for preparing teaching data, MLIT have started the working group of Industry-academia-government collaboration and are proceeding with discussions. At present, teaching data for cracks in bridges and tunnels has been prepared. MLIT is going to realize future advanced inspection, such as accumulating and using high-definition photographs with positional information taken by robots and inspection records in a 3D model. This will be possible to grasp the secular change of damage and deformation easily.

Keywords: AI, Inspection; Teaching data; Drone; Bridge; Tunnel