Firefighters Utilizing AI to Locate Wildfires
California firefighters are using artificial intelligence to help spot wildfires, feeding video from more than 1,000 cameras strategically placed across the state into a machine that alerts first responders when to mobilize.
As an example of the potential of the ALERTCalifornia AI program launched last month, the camera detected a fire that broke out at 3 a.m. local time in the remote, lush Cleveland National Forest about 80 kilometers east of San Diego.
As people slept and darkness covered the smoke, it could have spread to a raging forest fire. But the AI alerted the fire captain, who called in about 60 firefighters, including seven engines, two bulldozers, two water tankers and two hand crews. The fire was out in 45 minutes, Cal Fire said.
Engineers at the University of California, San Diego have developed DigitalPath, a company based in Chico, California, using artificial intelligence. The platform is based on 1,038 cameras installed by various public agencies and utilities across the state, each capable of rotating 360 degrees. command of remote operators.
After the AI program began on July 10, Cal Fire offered other examples of AI alerting fire captains before making a 911 call, though it did not yet have a comprehensive report.
UCSD geology and geophysics professor Neal Driscoll and ALERTCalifornia’s principal investigator said the sample size is too small to draw any conclusions yet.
Cal Fire hopes the technology can one day serve as a model for other states and countries around the world. This need is underscored by unusually destructive wildfires in Hawaii, Canada and the Mediterranean this season.
“It’s 100 percent applicable all over the world, especially now that we have much larger and more widespread fire systems and climate change,” said Suzann Leininger, a Cal Fire intelligence specialist in El Cajon, east of San Diego. .
Part of Leininger’s job is to help the machine learn. He checks the camera network for a previously recorded video of what the AI thinks is a fire, and then tells the machine if it was correct with a binary yes or no response. Any phenomenon can trigger a false positive: clouds, dust, even a smoky truck.
With hundreds of experts repeating the exercise up and down the state, the AI has already become more accurate in just a few weeks, Driscoll said.
In addition to the network of cameras, the platform will gather vast amounts of additional data, including an aerial survey to determine the vegetation that would feed future fires and map the ground surface beneath the canopy, Driscoll said.
Aircraft and drones also collect infrared and other wavelength data beyond human vision.
In winter, the platform can measure atmospheric rivers and snow cover. The UCSD team is also collecting data on the burns and their effects on erosion, sediment dispersal, water quality and soil quality, Driscoll said.
The information, available to any private company or academic researcher, could eventually be used to model fire behavior and improve as-yet-unexpected artificial intelligence applications for studying the environment.
“We’re in an extreme climate now. So we’re giving them the information because this problem is bigger than all of us,” Driscoll said. “We need to use technology to help move the needle, even if it’s a little bit.”
