2023 Fiscal Year Final Research Report
3D Material Distribution Mapping by Acoustic Camera in Extreme Underwater Environment
| Project/Area Number |
20K19898
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61050:Intelligent robotics-related
|
|---|
| Research Institution | Japan Advanced Institute of Science and Technology |
|---|
Principal Investigator |
JI YONGHOON 北陸先端科学技術大学院大学, 先端科学技術研究科, 准教授 (90823766)
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Keywords | 音響カメラ / 水中ロボット / SLAM / 極限環境 / マテリアル分布 |
|---|
| Outline of Final Research Achievements |
An acoustic camera which is mounted on an underwater robot can measure 3D space even in turbid water when exploring the underwater environment.
This study aims to establish a method to generate an underwater material distribution map that includes not only dense 3D shape information but also material information in the environment from the acoustic camera, in order to enable an intuitive understanding of the surrounding environment.
We implemented a system that generates an underwater material distribution map by establishing novel methodologies to recognize the materials of objects captured by an acoustic camera using semantic segmentation and generate a 3D map including the recognized material information.
|
| Free Research Field |
知能ロボティクス
|
| Academic Significance and Societal Importance of the Research Achievements |
新空港，港湾，海底トンネルの建設及び干拓事業などの水辺の開発はもちろん，福島第一原子力発電所の原子炉のような人間が入れない環境に対する調査活動での遠隔操作による水中ロボットの活用に注目が集まっている．本研究で実現した音響カメラの計測情報から周囲環境の密な3次元マテリアル分布マップを復元する技術は，周囲環境の直感的な認識が不可欠な無人水中ロボットの運用において極めて重要であり，提案技術の成果はさらに将来的にも水中探査や作業の自動化において大きな技術革新をもたらすと期待され，本研究の学術的な意義が大きい．
|
