2012年2月21日（火）18:00-20:00 吹田キャンパス 工学部 U2-211にて第44回創成塾を開催いたします。
18:00-18:20 Stefan Kopp （Bielefeld University）
18:30-18:50 鄭 寛浩 （大阪大学，基礎工学研究科）
講演1：Stefan Kopp （Bielefeld University）
“Talking hands — what gestures do in human dialogue and can do for robots”
Nonverbal behaviors are an integral part of natural human communication and collaboration, e.g. to complement speech, to establish joint attention, to regulate dialogue, or to express emotions. As we strive for natural communication also between humans and robots, this implies that every behavior shown or not shown by a robot may be taken as being indicative of some internal state or intended message of the system. We thus need careful models for the multimodal behavior our systems exhibit. From all nonverbal behavior, co-verbal gesturing stands out as it fulfills multiple functions alongside verbal communication and, to this end, is closely tied to speech in time, form, meaning, and pragmatics. I will present results from our work on understanding and modeling coordinated speech and gesture. This will include findings from a fine-grained empirical analysis on human co-verbal gestures in dialogue, a cognitively inspired computational model for speech and gesture generation, and results from evaluation experiments carried out with several virtual agents as well as a humanoid robot.
講演2：鄭 寛浩 （大阪大学，基礎工学研究科）
“How Many Social Robots Can One Operator Control?”
This study explores the nature of the multi-robot control problem for social robots. It begins by modeling the overall structure of a human-robot team for social interactions, and implements it for specific application to dialog-based interactions. Operator activity during control of a social robot is studied. Customer satisfaction is proposed as an important metric for evaluating the performance of a human-robot team for social interactions with customers. Based on the modeling, fan-out of a social robot team can be calculated, and the performance of the team is estimated by simulation. A field trial was conducted in a shopping mall to demonstrate a successful deployment of social robots for a real-world application with ensured performance prior to installation using our modeling and simulation approach.