Group dynamics play out in VR
Technology Research News
Shared virtual environments are often fantasy
spaces where people interact using avatars endowed with supernatural powers.
In contrast, simulations of large, real places are usually dedicated to
traffic flows or disasters and involve independent software agents rather
than avatars controlled by humans.
A researcher at Kyoto University in Japan has combined the two
ideas to make a virtual environment that allows large numbers of people
and software agents to interact in a space that closely models a real
place. The software, dubbed FreeWalk, allows programmers to model virtual
environments or real spaces and supports large groups of avatars and agents.
The work taps social psychological methods, multi-agent simulations
and multi-user environments, said Hideyuki Nakanishi, a research associate
at Kyoto University. Developers writing applications for FreeWalk can
construct realistic spaces by mapping photographs of a real space onto
the surfaces of a three-dimensional model of that space. Nakanishi has
modeled a shopping district in Kyoto, complete with street and subway
signs and store window displays.
Avatars and software agents appear as humans, and both human users
and agents use social cues like lip movement and gestures to inform their
interactions with each other. Agents respond to the social cues of human
users in order to mediate conversations and participate in group behavior,
People tend to respond to spaces and people on computer screens
as they respond to real spaces and people, said Nakanishi. FreeWalk takes
advantage of this tendency by making the virtual environment believable
enough that people behave appropriately for the context.
Nakanishi has built a demo system for conducting virtual, or telecommute,
evacuation drills. The system mimics disaster situations that include
real social interactions, and can eliminate the cost and disruption of
physically moving large numbers of people, said Nakanishi.
The system uses speech recognition software to convert a user's
words to text, and speech synthesis software to produce vocalizations
for agents and avatars. Keyword matching provides agents with a rudimentary
form of language understanding. The speech recognition and speech synthesis
engines run on the user's computer, which minimizes the amount of data
transmitted between users.
The server running the simulation communicates with client systems
only to add and delete avatars and agents. Voice, posture and gesture
cues are transmitted directly between clients in a peer-to-peer configuration.
Agents and avatars are programmed with basic behaviors like walking
and avoiding collisions with each other, allowing programmers to concentrate
on high-level behavior, according to Nakanishi.
FreeWalk is in its third iteration. The original program, developed
in 1996, was a spatial videoconferencing system. The second iteration
added software agents that facilitated conversations between people by
asking questions when the conversation faltered. The current version of
FreeWalk adds a simulated place that provides a context for group behavior
with agents as extras in the scene.
Nakanishi's next step is to add details like fire and smoke. "I
am aiming for the complete simulation of our living space," he said.
To make the simulations more realistic and accurate, real-world
human behaviors must be recorded, analyzed and modeled, said Nakanishi.
Practical applications that use the simulation are possible in five to
ten years, he said.
The work appeared in the April, 2004 issue of the International
Journal of Human-Computer Studies. The research was funded by the
Japan Science and Technology Agency (JST). FreeWalk is open source software
available at www.lab7.kuis.kyoto-u.ac.jp/freewalk.
Timeline: 5-10 years
TRN Categories: Data Representation and Simulation; Human-Computer
Interaction; Computers and Society
Story Type: News
Related Elements: Technical paper, "FreeWalk: a Social Interaction
Platform for Group Behavior in a Virtual Space," International Journal
of Human-Computer Studies, April, 2004.
May 19/26, 2004
Solar crystals get 2-for-1
Evolution trains robot
Group dynamics play
out in VR
could cool chips
turns plastic green
Research News Roundup
Research Watch blog
View from the High Ground Q&A
How It Works
News | Blog
Buy an ad link