| Project/Area Number |
18K18134
|
|---|
| Research Category |
Grant-in-Aid for Early-Career Scientists
|
| Allocation Type | Multi-year Fund |
|---|
| Review Section |
Basic Section 61050:Intelligent robotics-related
|
|---|
| Research Institution | Ritsumeikan University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2023-03-31
|
| Project Status |
Completed (Fiscal Year 2022)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2020: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2019: ¥1,820,000 (Direct Cost: ¥1,400,000、Indirect Cost: ¥420,000)
Fiscal Year 2018: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 転移学習 / 記号創発ロボティクス / 生成モデル / 生活支援ロボット / 概念獲得 / カテゴリゼーション / ベイズ推論 / 教師なし学習 / 概念形成 / カテゴリ / 概念 / 記号創発 / マルチモーダル / 認知モデル |
|---|
| Outline of Final Research Achievements |
This research aimed to construct a computational model that acquires prototypes (typicality) in concepts, enabling transfer learning based on the spatial concepts by utilizing a probabilistic generative model as the foundation to acquire spatial concepts in a bottom-up manner from multimodal information such as images, positions, and language. We conducted experiments on transfer learning and prototype acquisition based on this model in a simulation environment, revealing that it is possible to predict location names similar to those of humans. Furthermore, we implemented daily life support tasks based on the acquired prototypes. Specifically, we realized a task where the robot brings unobserved objects by leveraging typical knowledge acquired through the constructed model, such as the fact that sponges are commonly observed in bathrooms.
|
| Academic Significance and Societal Importance of the Research Achievements |
人間の認知における概念は，連続的な実世界の離散化を可能とし，言語によるコミュニケーションと深く関係している．概念が感覚運動情報から獲得されるメカニズムの解明は，認知発達ロボティクスにおける重要な挑戦である．本研究は，構成論的アプローチに基づき，概念の転移学習とプロトタイプ（典型）獲得のメカニズムを計算論モデルにより構築し，これを可能とする数理モデルを明らかにした．さらに，この計算論モデルを生活支援ロボットに応用し，人間との言語的なコミュニケーションを通じて現場環境で役に立つ仕事を実現した．この研究の成果は，国際的な知能ロボティクス競技会や国際会議において，人工知能学会賞等の多数の賞を受賞した．
|
