Spontaneous Avatar Behavior for Human Territoriality

Abstract

This paper presents a new approach for generating believable social behavior in avatars. The focus is on human territorial behaviors during social interactions, such as during conversations and gatherings. Driven by theories on human territoriality, we define a reactive framework which allows avatar group dynamics during social interaction. We model the territorial dynamics of social interactions as a set of social norms which constrain the avatar's reactive motion by running a set of behaviors which blend together. The resulting social group behavior appears relatively robust, but perhaps more importantly, it starts to bring a new sense of relevance and continuity to virtual bodies that often get left behind when social situations are simulated. We carried out an evaluation of the technology and the result confirms the validity of our approach.

Prediction in social path following

“…We implemented our approach in our CADIA Populus social simulation platform [Pedica and Vilhjálmsson 2010], resulting in a series of short prototype scenarios. The platform is based on the perception-reaction model, it provides human behaviors and faceto-face interaction including typical gestures, animations and territorial dynamics.…”

“…Actually this last behavior is made even a bit more interesting and complex by the social simulation platform, which drives the conversation group. It executes a common attention rule in the group: if one of the conversants sees someone approaching, it triggers the others to also look at the agent [Pedica and Vilhjálmsson 2010]. …”

More realistic, flexible, and expressive social crowds using transactional analysis

“…Previous research has pointed out that when agents socialize there is a mutually understood center of the conversation, called the formation nucleus (see Scheflen and Ashcraft [30] and Pedica and Vilhjálmsson [25]). When agents socialize they usually are within a small radius of this formation nucleus.…”

“…This is achieved by the creation of proxy agents, or spaces that are considered occupied by the crowd simulation engine but which are not rendered to the screen. Pedica et al [24,25] proposed an algorithm for agents talking in groups based on human territoriality theory. Carstendottir et al [6] focus on where agents in a crowd will choose to sit in a cafe or restaurant.…”

Lifelike interactive characters with behavior trees for social territorial intelligence

“…The gaming industry is slowly evolving the art of story telling but no matter how compelling the graphics or thrilling a story, awkward character behavior often breaks player immersion. In previous seminal work [Pedica and H. Vilhjálmsson 2010] we showed how the social theories of human territoriality and face-to-face interaction can serve as a solid base to model reactions expected by users when interacting with virtual characters. We have now integrated our reactive approach for social territoriality with Behavior Trees (BTs), an emerging game A.I.…”