Self-organizational imitation learning process through unsegmented human-robot interaction
| Project/Area Number |
20800060
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (Start-up)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Perception information processing/Intelligent robotics
|
|---|
| Research Institution | Ritsumeikan University |
|---|
Principal Investigator |
TANIGUCHI Tadahiro Ritsumeikan University, 情報理工学部, 助教 (80512251)
|
|---|
| Project Period (FY) |
2008 – 2009
|
| Project Status |
Completed (Fiscal Year 2009)
|
| Budget Amount *help |
¥3,289,000 (Direct Cost: ¥2,530,000、Indirect Cost: ¥759,000)
Fiscal Year 2009: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2008: ¥1,729,000 (Direct Cost: ¥1,330,000、Indirect Cost: ¥399,000)
|
|---|
| Keywords | 模倣学習 / 分節化 / 自己組織化型学習 / 時系列 / 統計的学習理論 / ベイズ / 情報論的学習理論 / 動作抽出 / 記号創発 / 機械学習 / 知能ロボット / パターン認識 / 自己組織化 / 自律適応系 / キーワード抽出 |
|---|
| Research Abstract |
In this research, we developed a machine learning method which enables a robot to obtain several unit motions from unsegmented human motion. In the method, the learning architecture reduces the dimensionality of observed time series data using singular vector decomposition, models the output data using Switching AR Model (SARM). After SARM encodes the data to code strings, a keyword extraction method extract unit words from the strings based on minimal description length criteria. The extracted words are revealed to be characteristic unit motions.
|
Report
(3 results)
Research Products
(34 results)
