| Project/Area Number |
07245103
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research on Priority Areas
|
| Allocation Type | Single-year Grants |
|---|
| Research Institution | Ritsumeikan University (1997)
The University of Tokyo (1995-1996) |
|---|
Principal Investigator |
ARIMOTO Suguru Ritsumeikan University, Faculty of Science and Engineering, Professor, 理工学部, 教授 (00029399)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KANEKO Makoto Hiroshima University, Faculty of Engineering, Professor, 工学部, 教授 (70224607)
UCHIYAMA Masaru Tohoku University, Graduate School of Engineering, Professor, 大学院・工学研究科, 教授 (30125504)
YOSHIKAWA Tsuneo Kyoto University, Graduate School of Engineering, Professor, 大学院・工学研究科, 教授 (60026177)
NAKAMURA Yoshihiko University of Tokyo, Graduate School of Engineering, Professor, 大学院・工学系研究科, 教授 (20159073)
|
|---|
| Project Period (FY) |
1995 – 1997
|
| Project Status |
Completed (Fiscal Year 1997)
|
| Budget Amount *help |
¥84,300,000 (Direct Cost: ¥84,300,000)
Fiscal Year 1997: ¥26,300,000 (Direct Cost: ¥26,300,000)
Fiscal Year 1996: ¥27,800,000 (Direct Cost: ¥27,800,000)
Fiscal Year 1995: ¥30,200,000 (Direct Cost: ¥30,200,000)
|
|---|
| Keywords | dexterity / robot / multi-fingered hand / bioengineering / impedance / reactive behavior / analysis of handwaor / assembly task / ロボットハンド / 把握 / パラレルロボット / インピーダンスマッチング / 学習 / 適応 / 技量の原理 / インピーダンス制御 / コンプライアンス制御 / 包み込み把握 / 探り動作 / 摩擦 / 重力フリーロボット / 協調動作 / 手先技量 / 作業計画 / 力学的理解 |
|---|
| Research Abstract |
The hand is called a branch of the brain. Manual dexterity is an intelligent human behavior generated unconsciously with sensory-motor organization in the central nervous system. Such intelligence is hidden from unconsciousness but it is an outcome of biological evolution. The purpose of this research is to understand the intelligence with expressing it by physical principles. The results of this research are summarized from the four viewpoints of 'modeling of tasks under geometrical constrains,' 'analysis of dexterity in tasks,' 'analysis of dexterity in living things' and 'synthesis of dexterity' as follows :
1. 'Modeling of tasks under geometrical constrains : 'We have found that the robot motion under geometrical constrains can be expressed by a circuit and have suggested that hybrid control of positions and forces can be analyzed as a DC system. A simulator of peg-in-hole insertion task have been developed in order for virtual reality to be implemented more naturally.
2. 'Analysis o
… More
f dexterity in tasks : 'We have analyzed dexterity in elementary tasks, that have been extracted from tasks with contact, after making measurement of their physical parameters. Our approach to the essence of dexterity has been achieved by evaluating even human intention in performing the task.
3. 'Analysis of dexterity in living things : 'We noted importance of impedance and started from measurement of impedance characteristics of human arms. We lighted the essence of dexterity from the standpoint of bioengineering after investigating usage and limits of impedance. We also studied extensively on measurement of abilities of multi-fingered hands in tasks,
4. 'Synthesis of dexterity : 'We have introduced a nonlinear structure that connects sensors to reactive behaviors and have abstracted the learning process of emergent motions as parameter learning. We have studied automatic planning of assembly tasks and have succeeded in planning of elementary tasks automatically on the basis of geometrical, mechanical and mathematical analysis. Also, we have constructed a unified control system of motion, force and compliance and have implemented it successfully by a fast parallel robot. Moreover, we have proposed a concept of natural motion for implementing more effective motion. Scale-dependent enveloping grasp by a multi-fingered hand has been realized. Less
|
