Journal of Japan Society for Fuzzy Theory and Intelligent Informatics
Online ISSN : 1881-7203
Print ISSN : 1347-7986
ISSN-L : 1347-7986
Original Papers
Emotional Motion Generation Using Emotion Representation Rules Modeled for Human Affection
Yoshikazu YANOAtsushi YAMAGUCHIShinji DOKIShigeru OKUMA
Author information
JOURNAL FREE ACCESS

2010 Volume 22 Issue 1 Pages 39-51

Details
Abstract
Non-verbal information is essential matter on communication between humans. Non-verbal representation for robot is also required as human could understand non-verbal sign within unintended expressions. Previous motion representation needs well-designed emotional motions prepared by designers in advances. As number of emotional motion has limitation according to huge burden for designers, and to be prepared the typical motion patterns with strong feelings, robots cannot bare adequate degree of emotion. In this paper, we proposed the emotional motion generation system using multilayer emotion model. Top and bottom layers of multilayer emotion model represent the emotion representation and mechanical properties correspondingly. Proposed middle layer represents the adjectival expressions which acts as emotion modification filters, such as “widely”, “rapidly”, “lower” and so on. Emotions are described using adjectival expressions as emotion expression rules which shows the relationships between motions and adjectival expressions. Emotion expression rules are obtained by organoleptic tests, which models for human affection of emotional motions. After 2-step experiments, emotion expression rules for motion generation were obtained and were applied to unintended motion for emotional motion generation. And we clarified the utility of proposed technique.
Content from these authors
© 2010 Japan Society for Fuzzy Theory and Intelligent Informatics
Previous article Next article
feedback
Top