Infrastructure for Multi-User, Multi-Spatial Collaborative Environments
Stephen Peters · 2001
The Problem: Current research into intelligent environments focuses on a single user interacting with a single space. While this is a useful interaction, it ignores the propensity of people to work on problems in teams, as well as the utility of moving projects from workspace to workspace. In this project, we extend an existing agent-based environment to seamlessly handle multiple users and spaces. Motivation and Previous Work: Recent research into intelligent environments [7, 3] has focused primarily on systems where a single user interacts with a single intelligent space, through the use of speech, gestures, or more mundane keyboard input. These systems often ignore handling multiple people, yet much interaction occurs in a conference or meeting room, with multiple people interacting with each other. In addition, these research efforts usually limit themselves to a single intelligent environment, ignoring the need for users to have their information and applications transferred with them astheymove. Attempts to achieve these capabilities without representing individuals as individuals per se, such as the KidsRoom project [1], creates systems that can only recognize users as aggregate concepts, and are limited to very simple feedback mechanisms, without the ability to target the feedback to one user. The EasyLiving project [2] incorporates the notion of individuals into its system, and cantrack multiple independent people within its room. However, it is still tied closely to one specific space.