June 6th, 2007
Using AI to train firefighters
Computer scientists at the University of Southern California (USC) have developed DEFACTO, a training program which uses artificial intelligence (AI) to help firefighters practice simulated emergency situations. The system is currently used by the Los Angeles Fire Department. DEFACTO has committees of AI ‘agents’ which can create disaster scenarios with images and maps seen in 3-D by the trainees. The software agents also evaluate the trainees’ answers and help them to take better decisions. As said one LAFD captain, ‘You can see if you’re heading toward a mistake much more quickly.’
On the picture above, you can see Los Angeles Fire Department Captain Ron Roemer using the DEFACTO training system (Credit: University of Southern California, Los Angeles). This system has been developed by Nathan Schurr, who just got his PhD, and other members of the group led by Milind Tambe, a professor at the Computer Science Department of USC.
Why these researchers decided to build this tool? “The idea was to speed up and make more useful a training simulation. The LAFD had previously run simulated training drills working by hand. [...] In one room, a B-team of veteran firefighters would make up a disaster, sending bulletins to another room, where trainees would decide where to send equipment and how to respond. [...] The system meant that dozens of top personnel would be tied up for hours.”
With DEFACTO, there is no need for a B-team. And trainees learned how to delegate some power to the AI software agents.
For more information, here is a link to Schurr’s disaster rescue simulations. The DEFACTO project is funded by the CREATE (Center for Risk and Economic Analysis of Terrorism Events), an interdisciplinary national research center based at the University of Southern California and funded by the Department of Homeland Security.
Finally, you might want to read a technical paper about this training system. “Using Multiagent Teams to Improve the Training of Incident Commanders” has been included in the 2006 Proceedings of the Industry Track of the Fifth International Joint Conference on Autonomous Agents and Multiagent Systems. Here is a link to this paper (PDF format, 8 pages, 801 KB).
Will this system be used by other fire departments in California or in the rest of the world? I don’t know yet, but the idea behind this project is promising.
Sources: USC Viterbi School of Engineering, May 25, 2007; and various websites
You’ll find related stories by following the links below.
