Developmental Skill Acquisition by Anthropomorphic robotic soft fingertip with randomly distributed receptors
| Project/Area Number |
16300056
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
HOSODA Koh Osaka University, Graduate School of Engineering, Associate Professor, 大学院工学研究科, 助教授 (10252610)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
ASADA Minoru Osaka University, Graduate School of Engineering, Professor, 大学院工学研究科, 教授 (60151031)
TAKAHASHI Yasutake Osaka University, Graduate School of Engineering, Research Associate, 大学院工学研究科, 助手 (90324798)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Project Status |
Completed (Fiscal Year 2006)
|
| Budget Amount *help |
¥13,000,000 (Direct Cost: ¥13,000,000)
Fiscal Year 2006: ¥3,600,000 (Direct Cost: ¥3,600,000)
Fiscal Year 2005: ¥3,800,000 (Direct Cost: ¥3,800,000)
Fiscal Year 2004: ¥5,600,000 (Direct Cost: ¥5,600,000)
|
|---|
| Keywords | artificial fingertip / tactile sensor / object manipulation / artificial skin / multi-modality / ひずみゲージ / フッ化ポリビニリデンフィルム / フィッシャーの線形判別 / こすりによる物体識別 / 分布型触覚 / ロボットハンド / 多層構造 |
|---|
| Research Abstract |
This research aimed to develop an anthropomorphic soft fingertip with randomly distributed receptors. The structure of the fingertip was similar to that of a human's ; it consisted of a bone, a body, a skin layer, and randomly distributed receptors inside. The fingertip was supposed to acquire meaningful information such as slippage and object texture from the outputs of receptors through experience.
1.An anthropomorphic soft fingertip was developed with randomly distributed receptors. A robot hand system was then developed utilizing these fingertips.
2.The ability of the fingertip was estimated by discriminating experiments on several types of texture.
3.A neural network is proposed to learn the relation between the tactile sensation obtained by the anthropomorphic fingertip and the vision. By the proposed network, the robot could begin to do the task by utilizing vision, gradually acquired relation between tactile and vision, and eventually, it could get robust and redundant expression on the slip.
4.Utilizing the network, the robot could perform a peg-in-hole task and a pick-up task. At the commencement, the task was done mainly by vision, but gradually the robot use tactile more. Eventually, the performance of the task grows because of utilizing tactile information.
|
Report
(4 results)
Research Products
(19 results)
