A Behavioral Animation System for Autonomous Actors personified by Emotions

This paper presents a system dedicated to producing behavioral animation involving autonomous actors. To increase their believability, this system not only allows defining their behavior but also contains an emotional model which generates emotions felt by the actors. These emotions are defined by emergence conditions describing how and when they should emerge. The reactions selected by the actors in response are defined by their character. The use of an emotional model increases the believability of virtual actors by preventing actors from reacting in same manner in identical contexts and gives the impression that each actor has its own distinct personality.