Surviving Jaws: Virtual Merman "Thinks" His Way To Safety
- Date:
- February 29, 2000
- Source:
- University Of Toronto
- Summary:
- Researchers at the Univerisity of Toronto have created self-animating virtual characters with functional bodies and brains that have behaviour, perception, learning and cognition centres.
- Share:
Using a mythical merman and hungry sharks, a University of Toronto computer science professor and two former graduate students have pushed the notion of artificial intelligence and virtual life to a new level.
In his creation of a virtual underwater world, Professor Demetri Terzopoulos has fashioned more than just a cool screen saver - he has given his animated characters the ability to think. A hungry shark circles ominously, looking for a nice meal, while a nervous merman searches for a place to hide. When the shark swims away, the merman dashes from behind large rocks to open water with the shark in hot pursuit. Will his cleverly devised plan allow him to reach safety or not?
"This is more than artificial intelligence," says Terzopoulos. "It's artificial life. Computer graphics, animation and virtual reality have advanced dramatically over the past decade. We are now able to create characters that are self-animating with functional bodies and brains that have behaviour, perception, learning and cognition centres."
Terzopoulos and his former students have developed the cognitive modelling language that enables animated characters to reason. For example, it enabled the virtual merman to formulate a plan of action by reasoning about his situation given certain knowledge, such as the fact that he cannot outrun sharks but can use underwater rocks to hide. "With cognitively empowered graphical characters, the animator need only specify a behaviour outline and, through reasoning, the character will automatically work out a detailed sequence of actions."
The potential for future applications are immense, Terzopoulos says. Cognitive modelling and the cognitive modelling language can become powerful tools for scientists, animators and game developers. His paper, co-authored with John Funge and Xiaoyuan Tu, was published at the 1999 ACM SIGGRAPH conference, the premier forum for research in computer graphics.
CONTACT:
Janet Wong
U of T Public Affairs
(416) 978-6974
jf.wong@utoronto.ca
Story Source:
Materials provided by University Of Toronto. Note: Content may be edited for style and length.
Cite This Page: