Presentation 2010-01-22
Robots That Learn to Communicate by Multimodal Interaction
Naoto IWAHASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents machine learning method L-Core that enables robots to learn to communicate linguistically from scratch through verbal and behavioral interaction with users. L-Core combines speech, visual, and tactile information obtained by interaction in the real world. It learns speech units, words, the concepts of objects, motions, grammar, and pragmatic and communicative capabilities, which are integrated in a dynamic graphical model. Experimental results show that through a practical, small number of learning episodes with a user, the robot was eventually able to understand even fragmental and ambiguous utterances, respond to them with confirmation questions and/or acting, generate directive utterances appropriate for the given situation, and answer questions. This paper discusses the importance of a developmental approach to realize natural situated human-robot conversation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Language / Robot / Learning / Communication / Motion / Multimodal
Paper # CQ2009-100,PRMU2009-199,SP2009-140,MVE2009-122
Date of Issue

Conference Information
Committee CQ
Conference Date 2010/1/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Communication Quality (CQ)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Robots That Learn to Communicate by Multimodal Interaction
Sub Title (in English)
Keyword(1) Language
Keyword(2) Robot
Keyword(3) Learning
Keyword(4) Communication
Keyword(5) Motion
Keyword(6) Multimodal
1st Author's Name Naoto IWAHASHI
1st Author's Affiliation National Institute of Information and Communications Technology()
Date 2010-01-22
Paper # CQ2009-100,PRMU2009-199,SP2009-140,MVE2009-122
Volume (vol) vol.109
Number (no) 373
Page pp.pp.-
#Pages 6
Date of Issue