2022 Fiscal Year Final Research Report
Research on Maintenance Support for Large Structures Using Knowledge-Based Point Cloud Processing
Project/Area Number |
20H02052
|
Research Category |
Grant-in-Aid for Scientific Research (B)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Review Section |
Basic Section 18030:Design engineering-related
|
Research Institution | The University of Electro-Communications |
Principal Investigator |
Masuda Hiroshi 電気通信大学, 大学院情報理工学研究科, 教授 (40302757)
|
Co-Investigator(Kenkyū-buntansha) |
遊佐 泰紀 電気通信大学, 大学院情報理工学研究科, 助教 (70756395)
|
Project Period (FY) |
2020-04-01 – 2023-03-31
|
Keywords | 点群処理 / 機械学習 / 設備保全 / 物体認識 / 形状モデリング / 劣化検出 / 3次元計測 / 構造解析 |
Outline of Final Research Achievements |
In recent years, the aging of large structures has become a major problem. In order to improve the efficiency of maintenance work, this research aims to investigate point cloud processing methods using machine learning and engineering knowledge. In this research, we developed five point processing methods for large-scale point clouds of enginnering facilities; (1) point cloud segmentation and object recognition methods using deep learning, (2) shape reconstruction methods from incomplete point clouds, (3) method for calculating optimal measurement positions for mobile robots,, (4) deterioration detection methods from point clouds using deep learning, and (5) structural analysis methods from incomplete point clouds.
|
Free Research Field |
設計工学
|
Academic Significance and Societal Importance of the Research Achievements |
本研究は,測量用のレーザスキャナで得られた点群を用いて,大規模な設備保全を効率的に行うための手法である.近年,機械学習が進歩しているが,点群の利用や設備保全への応用においては,必ずしも有効な手法とはなっていない.本研究では,機械学習を工業設備の保全に利用するために,5つの課題を設定して点群処理手法を開発し,その有効性を検証している.本研究は,工学的に新しい手法を提案するとともに,実際の大規模点群にも活用できるという点で実用的にも有用なものである.
|