USC researchers use artificial intelligence to predict wildfire activity

Artificial intelligence helps USC researchers predict wildfire activity

SACRAMENTO — In California, wildland firefighters need all the help they can get, and time is something that can't be wasted. One university is studying how to use artificial intelligence to predict a wildfire's next move.

Year after year, wildfires have ravaged California.

Assad Oberai isn't a firefighter. He's a professor of aerospace and mechanical engineering at the University of Southern California, but he thought he and his team could help. He said that computer models already exist to try and predict fire movement.

"The problem is the things that go into these tools, like terrain, the fuel map, what are the wind conditions—these things are very uncertain," Oberai said.

So what his team did was gather historical wildfire data from high-resolution satellite images and add in a little artificial intelligence.

"Now we have a better estimate of where the wildfire is at, how much it has burned, how much the atmosphere has been heated up by the wildfire," Oberia said "And because of that, we can then make an accurate prediction of where the wildfire is going to go."

Oberai said that this is not currently being used in the field. They still have a ways to go, but the hope is that once it's ready it will save firefighters precious moments that can save homes and, more importantly, lives.

A number of organizations are currently working to integrate artificial intelligence to help fight fires. They hope it might be ready for public use in the next few years.

Read more
f

We and our partners use cookies to understand how you use our site, improve your experience and serve you personalized content and advertising. Read about how we use cookies in our cookie policy and how you can control them by clicking Manage Settings. By continuing to use this site, you accept these cookies.