• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2018 Fiscal Year Final Research Report

Comprehensive realization of imitation related tasks of developmental scales based on partially observable Markov decision process theory

Research Project

  • PDF
Project/Area Number 15K00341
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Research Field Soft computing
Research InstitutionSaga University

Principal Investigator

ITOH Hideaki  佐賀大学, 理工学部, 准教授 (20345375)

Project Period (FY) 2015-04-01 – 2019-03-31
Keywords確率的情報処理 / POMDP / 多機能エージェント / 発達尺度 / 模倣 / POMCP / 適応制御 / 最適制御
Outline of Final Research Achievements

In this study, we have developed several systems to perform imitation related tasks of developmental scales such as whole body movement imitation and figure copying. We have developed a system in which the agent can determine through dialogue what kind of imitation should be performed. We have also developed a system in which the agent itself can automatically optimize what kind of information processing should be performed in doing each imitation task. We realized the dialogue and information processing optimization in a unified manner using the partially observable Markov decision process (POMDP) theory.

Free Research Field

ソフトコンピューティング,人工知能

Academic Significance and Societal Importance of the Research Achievements

本研究では、模倣関連タスクを遂行するための各種システムを開発し、どのような模倣を行うべきかをエージェントが対話から判断することや、各タスクの遂行においてどのような情報処理を行えばよいかをエージェント自身が自動的に最適化することを、部分観測マルコフ決定過程(POMDP)理論という汎用的な理論基盤の上で統一的に実現することができた。これは今後の多機能エージェントの研究・開発の基礎を築いた、意義のある成果であると考える。

URL: 

Published: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi