Systematization of Drone Audition Capable of Listening to Sounds in Real World Environments

2021 Fiscal Year Final Research Report

• Project/Area Number: 19H00750
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Medium-sized Section 20:Mechanical dynamics, robotics, and related fields
• Research Institution: Waseda University
• Principal Investigator: Okuno Hiroshi G 早稲田大学, 次世代ロボット研究機構, その他(招聘研究員) (60318201)
• Co-Investigator(Kenkyū-buntansha): 公文 誠 熊本大学, 大学院先端科学研究部(工), 教授 (70332864) ; 干場 功太郎 東京工業大学, 工学院, 助教 (50782182) ; 鈴木 太郎 千葉工業大学, 未来ロボット技術研究センター, 上席研究員 (80710368) ; 劉 浩 千葉大学, 大学院工学研究院, 教授 (40303698)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: ドローン聴覚 / ロボット聴覚 / カイトプレーン / 凧式ドローン / 生物規範型静音翼 / 三次元点生成 / 複数GNSS / 認知ドローン聴覚

Outline of Final Research Achievements

In "drone audition," which aims at acoustic monitoring from the sky by drones, dealing with ego-noise caused by rotors and airflow is much more important than in robot hearing. In this research, the performance of acoustic processing was improved by selecting frequency bands according to noise characteristics; silent propellers designed based on bio-norms and fluid dynamics were found to be not only quieter but also effective in reducing specific frequency bands of ego-noise; the acoustic characteristics was exploited to improve sound source location; a high-precision 3D position-and-pose estimation methods with a single GNSS antenna robust against the instability of drones was developed for constructing 3D point clouds; drive-and-glide flight navigation for a kiteplane was proposed; a method to design a cognitive drone hearing system based on the perceptual behavior cycle was proposed. These wide-spectrum outcomes have established the system of drone audition.

Free Research Field

ロボット聴覚

Academic Significance and Societal Importance of the Research Achievements

【学術的意義】①空からの複数音源定位に対する総合的手法，②プロペラの音響特性を活用したロバストな高速音源定位法，②CFDと生物規範型静音プロペラの開発を通じた静音化と音響処理との結合，③カイトプレーンによる滑空時の静音化を活用した音源定位技術，④機体の揺れに頑健な位置姿勢推定法と3D点群構築法，を通じてドローン聴覚の要素技術を体系化し，認知ドローン聴覚への展開を試みた．
【社会的意義】　ドローンの騒音問題に対して静音ドローン（Quiet Drones）が欧米で注目をされ始めており，本研究が取り組んだ静音ドローン技術は，健全なドローンの普及に寄与すると期待される．