A two-layered approach to communicative artifacts

  • Authors:

  • Yong Xu;Tatsuya Hiramatsu;Kateryna Tarasenko;Toyoaki Nishida;Yoshiyasu Ogasawara;Takashi Tajima;Makoto Hatakeyama;Masashi Okamoto;Yukiko I. Nakano

  • Affiliations:

  • Kyoto University, Graduate School of Informatics, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Kyoto, Japan;Kyoto University, Graduate School of Informatics, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Kyoto, Japan;Kyoto University, Graduate School of Informatics, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Kyoto, Japan;Kyoto University, Graduate School of Informatics, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Kyoto, Japan;The University of Tokyo, Graduate School of Information Science and Technology, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Tokyo, Japan;The University of Tokyo, Graduate School of Information Science and Technology, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Tokyo, Japan;The University of Tokyo, Graduate School of Information Science and Technology, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Tokyo, Japan;Tokyo University of Technology, Katayanagi Advanced Research Laboratories, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Tokyo, Japan;Tokyo University of Agriculture and Technology, Department of Computer and Information Sciences, Yoshida-Honmachi, Sakyo-Ku, 606-8501, Tokyo, Japan

  • Venue:
  • AI & Society
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

A key issue in social intelligence design is the realization of artifacts that can fluently communicate with people. Thus, we proposed a two-layered approach to enhance a robot’s capacity of involvement and engagement. The upper layer flexibly controls social interaction by dynamic Bayesian networks (DBN) representing social interaction patterns. The lower layer improves the robustness of the system by detecting rhythmic and repetitive gestures. We designed a listener robot that can follow and record humans’ explanation on how to assemble and/or disassemble a bicycle. The implementation of this system is described by assembling the key algorithms presented in this paper.