Capturing Behaviour for the Use of Avatars in Virtual Environments THOMAS E. WHALEN, Ph.D., 1 DORINA C. PETRIU, Ph.D., 2 LUCY YANG, Ph.D., 3 EMIL M. PETRIU, Ph.D., 3 and MARIUS D. CORDEA, M.A.Sc. 3 ABSTRACT Avatars, representations of people in virtual environments, are subject to human control. However, for most applications, it is impractical for a person to directly control each joint in a complex avatar. Rather, people must be allowed to specify complex behaviours with simple instructions and the avatar permitted to select the correct movements in sequence to execute the instruction. This requires a variety of technologies that are currently available. Human behaviour must be captured and stored it so that it can be retrieved at a later time for use by the avatar. This has been done successfully with a variety of haptic interfaces, with visual ob- servation of human head movements, and with verbal behaviourin natural language applica- tions. The behaviour must be broken into atomic actions that can be sequenced with a regular grammar, and an appropriate grammar developed. Finally, a user interface must be developed so that a person can deliver instructions to the avatar. 537 CYBERPSYCHOLOGY &BEHAVIOR Volume 6, Number 5, 2003 © Mary Ann Liebert, Inc. INTRODUCTION T HE INHABITANTS OF VIRTUAL ENVIRONMENTS can be classified as bots and avatars.A bot is an autono- mous agent that pursues its own goals. In contrast, an avatar—a representation of a human being—is under the direct control of that human being. For example, virtual meetings may consist of a repre- sentation of a conference room inhabited by hu- manoid figures that represent the participants. It may also contain a humanoid bot that performs sec- retarial tasks, such as fetching documents from a file cabinet as required. In practice, the distinction between bots and avatars is not as clear-cut as it might appear. On one hand, bots are designed by people and are given both overall goals and the means to achieve those goals at the time that they are designed. In a sense, they remain under the general control of their de- signers, though at a distant remove. On the other hand, it is not practical to control an avatar in real time at the most detailed level. Avatars consist of a number of pieces (segments) that are connected at joints. All movement consists of the rotation of a segment about an axis through the joint. A typical humanoid avatar like those defined by the H-Anim Standard 1 contains more than four dozen joints (not including the additional joints in the spine which have limited mobility). A person cannot specify the rotation of each of forty-eight joints in real time ex- cept by wearing a full-body sensory suit and acting out the motions. This is practical for computer- animated movies but not for most other applica- tions of virtual reality. Instead, avatars borrow some of the technology of bots so that the person controlling an avatar need only specify high-level behaviours (“sit in the chair,” “move closer to Bob,” “turn on the projector”) and 1 Communications Research Centre, Ottawa, Ontario, Canada. 2 Carleton University, Ottawa, Ontario, Canada. 3 University of Ottawa, Ottawa, Ontario, Canada.