Modeling the Personal Space of Virtual Agents for Behavior Simulation

Abstract

Abstract-In this paper we propose a mathematical model for the concept of Personal Space (PS) and apply it to simulate the non-verbal communication between agents in virtual worlds. The distance between two persons reflects the type of their relationship. Human-like autonomous virtual agents should be equipped with such capability to simulate natural interactions. We define three types of relationships; (1) stranger relationship, (2) business relationship, and (3) friendly relationship. First we model the space around an agent as a probability distribution function which reflects at each point in the space the importance of that point to the agent. The agent updates dynamically this function according to (1) his relation with the other agent, (2) his face orientation, and (3) the evolution of the relationship over time as a stranger agent may become a friend. We demonstrate the concept on a multiagent platform and show that space-aware agents exhibit better natural behavior.

The Impact of a Virtual Agent's Non-Verbal Emotional Expression on a User's Personal Space Preferences

“…As an essential aspect in social interactions, PS is a recurring subject of investigations. Being a non-verbal communication channel [4], human proxemic behavior, defined as the use of space in interpersonal interactions [15], reflects the nature of a relationship between an individual and its interactants. Thus, PS is considered as a personal safety zone humans try to maintain around themselves.…”

“…Research indicated that the exact size and shape of an individual's PS depends on numerous factors [16]. Various personal and social characteristics, such as gender, age, culture, or affective expressions [4,20,40] impact the PS preferences as well as environmental factors, such as lighting and in-or outdoor location [1,12]. Often found are elliptically-shaped PS zones, with roughly twice as much space in the individual's front compared to its back and side areas [4].…”

Emotional virtual crowd on task completion in virtual markets

“…Durupinar et al combined OCEAN personality model, pleasure‐arousal‐dominance model, and Ortony, Clore, and Collins model in crowd simulation. The modeling of the personal space also improves the social presence of virtual agents …”

An anthropomorphic navigation scheme for dynamic scenarios

“…Amaoka et al in [13] propose a mathematical model of personal space in order to apply it to agents in virtual worlds and simulate realistic behavior. They use two 2dimensional Gaussian functions which represent the personal space around an individual.…”