Can startup’s firefighting drones save us from another fire catastrophe?
With flame-ravaged Bay Area communities still mired in a tough recovery after California’s worst fire season destroyed more than 1,000 Bay Area homes, a Silicon Valley startup says its artificially intelligent firefighting drones could help stop future catastrophes.
If drones from Rain Industries had been in position around the Bay Area during this August’s lightning storms, the aircraft could have contained 72% of the fires within 10 minutes of ignition, the Palo Alto firm’s co-founder and CEO Maxwell Brodie said. “This is a transformative technology,” Brodie said. “If it is us or someone else that does this, it doesn’t really matter. This will happen.”
After starting out with a smaller, six-rotor prototype drone that successfully doused small fires by dropping balls full of retardant, Rain is now testing autonomous aircraft resembling small helicopters that it says can fly preemptively during potentially hazardous wildfire conditions and use their infrared sensors to locate and combat flames when they first erupt. Alternatively, the drones could take to the air as soon as flames are detected by the hundreds of fire-spotting cameras already positioned throughout California, existing lightning-strike-detection antennas or weather satellites.
“We are going after and solving the rapid response piece,” Brodie said.
Rain’s technology has already drawn interest from area fire officials, including Dixon fire chief Todd McNeal, who had used drones as an “eye in the sky” when he was fire chief in Twain Harte in the Sierra Foothills east of Stockton.
“By no means will they or I ever tell somebody that this will replace firefighters on the ground,” McNeal said. “We’re not going to full autonomy and sit back and let the machines do the work. The intent is to speed up the detection, to speed up the initial response, and deliver some sort of suppression agent to hold it, to slow it down.”
Fires in the western U.S. are regularly setting new records for size and damage, creating an urgent need for new tactics and technologies, McNeal said. Rain’s drones can help “fill in the coverage map” in areas further from fire departments and firefighting aircraft, he said.
The “Mark 2” drones Rain is now testing can carry up to 400 pounds of liquid retardant to extinguish fires or create containment lines that prevent flames from spreading, Brodie said. The company hopes to sell service contracts to Cal Fire and local fire departments, as well as insurance companies and utilities like PG&E that have suffered multi-billion-dollar wildfire losses in recent years.
Rain’s system could be ready to go in a year, Brodie said.
Each drone would be parked inside a secure shelter and launched in response to fire-watch cameras or satellite imagery. It would navigate autonomously to the ignition site using “computer vision,” its infrared sensors feeding data to AI software to estimate the size, growth rate, and direction of the fire, before “deciding” whether to drop retardant on the fire, create a firebreak ahead of it, or summon more drones. Human operators would oversee and approve its operations.
“Drones have certainly come into the forefront in the last several years,” said former Menlo Park Fire Protection District director Bart Spencer, who now works for Rain. “It makes sense that you take something that’s on the cutting edge of technology and you’re adding a firefighting skill to that. You’re getting a resource to a place that oftentimes is not reachable or not reachable in a reasonable amount of time.”
But if Rain is to successfully deploy its drones, it will have to surmount several challenges, including “accurately identifying the sheer number of hot spots in a large and fast-expanding fire,” said UC Berkeley forestry expert William Stewart. Piloted helicopters can accurately drop much larger volumes of retardant, said Stewart, who also cited “the airspace challenge of drones interfering with much larger, and much more effective, helicopters” and potential problems with wildfire-related turbulence.
Brodie acknowledged that high winds are a challenge for drones, and said bigger aircraft tend to handle them better. Rain will test the Mark 2 to determine its wind tolerance, he said. “If we need to go larger than that it’s not fundamentally a problem as far as the technology goes,” Brodie said. “With larger vehicles you need fewer vehicles in an area.”