2015 Fiscal Year Final Research Report
Active motion-planning and generation based on Learning-from-Observation using audiovisual
| Project/Area Number |
23240026
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
Nakaoka Shin'ichiro 国立研究開発法人産業技術総合研究所, 知能システム研究部門, 主任研究員 (60443206)
Kudoh Shunsuke 電気通信大学, 大学院情報システム学研究科, 准教授 (90582338)
|
|---|
| Project Period (FY) |
2011-04-01 – 2016-03-31
|
| Keywords | 知能ロボティクス |
|---|
| Outline of Final Research Achievements |
This project established a motion generation method for dancing robots from observation. The method divides actions into two classes: tasks, common actions to describe what-to-do and skills, various variations of actions due to dancers and performances to represent how-to-do. The method proposes to utilize Labanotation to describe tasks, and to establish parameters for skill representations. Through these task and skill modeling, the method succeeds 1) to performance of various folk dances by humanoid robots, 2) to represent various personal and occasional differences of performances. Furthermore, on top of these scientific theories, the project successfully classifies folk dances of Taiwanese indigenous people and identifies the classification tree of tribes based on those folk dances has a high correlation with the one based on their social institutes. The PI believes this is one of the promising direction to create a new scientific discipline to combine engineering and humanities.
|
| Free Research Field |
コンピュータビジョン
|
