Development of a Trans-Field Science Learning System with a Humanoid Robot as an Interface

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K00912
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Science education
• Research Institution: Tokyo Gakugei University
• Principal Investigator: MATSUURA Shu 東京学芸大学, 教育学部, 教授 (70238955)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: e-Learning / ヒューマノイド・ロボット / Topic Maps / 対話プラットフォーム / 対話生成 / ヒューマン・インタフェース / オンライン学習システム / ヒューマノイドNAO

Outline of Final Research Achievements

A Topic Maps-based online learning system with the backbone of structured knowledge was constructed, and a dialog system of a humanoid robot NAO with a direct connection with the learning system was developed as an information interface. A three-step model for the dialog generation was created for NAO. To query the server, the modules for various association types were used by the call of predetermined key-phrases that the humans utter. Also, a focus-interest model was created to construct the temporary memory for speech recognition. Further, a dialog frame was designed for the effective demonstration of the collaborative human-humanoid dialog, based on an activation-core-feedback sequence. The proposed system has been applied to the classrooms of a university and an elementary school.

Free Research Field

科学教育