Maxwell Brody vividly remembers the devastating wildfires he experienced as a child growing up in the interior of British Columbia.
One night in 2003, lightning struck a tree around 4 a.m., starting a massive fire that destroyed Okanagan Mountain Park. The winds picked up, the sky turned orange, and more than 30,000 people were evacuated from their homes. Brody remembers helping his father install a soaker hose to protect the cedar roof from falling ash.
That experience inspired Brody nearly two decades later to launch a software startup that gives autonomous helicopters and other aircraft the ability to sense and suppress wildfires.
“That’s something you never forget as a kid,” said Brody, co-founder and CEO of Alameda-based company Reign. “As these fires become more frequent and severe, it becomes even more important that we expand our response capabilities, including being able to respond at night, in smoky conditions, and in high winds.”
Brody is part of a small but growing group of California entrepreneurs who promise new technologies, many powered by artificial intelligence, that could dramatically change the way firefighters prevent and extinguish wildfires. He is one of the executives.
Facing budget shortfalls, fire departments have traditionally been wary of adopting expensive and experimental firefighting techniques that have not been proven in the field. But the unprecedented scale of the Los Angeles fires, which destroyed thousands of buildings and killed at least 27 people, has brought renewed interest and urgency to finding more effective ways to fight wildfires. .
“It’s a whole different scale…We have to come up with new ways to fight (fires),” said Josh Wilkins, a former fire chief with the San Bernardino County Fire Department.
In Silicon Valley, big tech companies like Google and AI giant Nvidia are investing in research that could help firefighters more effectively spot and track wildfires.
In 2021, Nvidia will partner with Lockheed Martin, the USDA Forest Service, and the Colorado Department of Fire Protection to help firefighters and incident commanders better understand how fires spread and We announced that we have created a digital version of the fire that will allow you to make suggestions. We will provide information to suppress it.
“The 21st century security technologies we are developing to respond to national security threats are unique to the complex environment of wildland fires,” said Dan Rhodan, senior program manager at the Lockheed Martin Artificial Intelligence Center in Connecticut. It can be directly applied to
AI-powered decision support devices may soon be able to support first responders’ command decisions, but that will depend on the availability of data and how close to real-time it is, Laudan said. .
Space agency NASA is also working on technology that would allow drones and remotely piloted helicopters to fly simultaneously to fight wildfires, even in low visibility conditions.
Fire departments across the state are already using AI tools run by the University of California, San Diego that can detect fires from video footage, allowing them to respond quickly to fires. The program, known as ALERTCalifornia, deploys more than 1,144 cameras and sensor arrays to capture live video around the clock.
The California Department of Forestry and Fire Protection collaborated with ALERTCalifornia and DigitalPath to develop the AI tool.
“This creates a network to watch over California,” said Cal Fire Chief David Acuña.
There are also some success stories. For example, last month, ALERTCalifornia’s AI system detected a fire in Black Star Canyon and alerted the Orange County Fire Department at 2 a.m. Firefighters extinguished the fire and contained the fire to less than a quarter acre.
Nevertheless, while ALERTCalifornia has helped save lives, the Los Angeles fires also exposed its limitations. The fire was fueled by strong winds and spread faster than firefighters could keep up.
To improve capabilities, Cal Fire is testing new equipment with BurnBot, a South San Francisco company that operates heavy vehicles that can perform controlled combustion with little or no smoke. The state-of-the-art vehicle, called the RX, is equipped with a propane torch that allows the operator to control the length and temperature of the flame. It also comes with a water spray nozzle and heavy roller for extinguishing fires.
Wilkins, who advises Burnbot and other wildfire prevention startups, believes the vehicles could have slowed the spread of the Los Angeles wildfires had they been deployed.
“Once you have a wind-driven fire, you’re battling embers,” Wilkins said. “Essentially, there are millions of matches flying through the air, and when one large bush catches fire, it sets off thousands of embers, and each one of those embers has the potential to ignite whatever it ignites. There is.”
Acuña said the agency is still evaluating Burnbot vehicles and waiting for data that will help determine how or if they will be used.
One of the obstacles to wider use of these futuristic firefighting tools, such as sensors that can detect smoke and fire, is a lack of private and public sector funding.
“This is a wake-up call for all of us about what we are up against,” said Sonia Kastner, co-founder and CEO of San Francisco-based Pano AI. spoke. “We need to fundamentally change the way we approach firefighting and natural resource management.”
Kaestner knows the challenges firsthand. She built an AI-powered platform to detect fires and alert emergency responders after the 2018 Camp Fire, which killed 85 people, burned 153,336 acres, and caused an estimated $16.5 billion in losses. We launched Pano AI.
Pano AI utilizes cameras placed at vantage points, such as cell phone towers, to scan the surrounding area and relay video images to emergency personnel. These are used in Ukiah and Rancho Palos Verdes, California, and other states.
The Department of Homeland Security operates a technology center within the Washington-based Science and Technology Directorate that has helped develop sensors to detect fires and toxic chemicals.
S&T and N5 Sensors conducted a controlled fire in Stafford, Virginia in advance of the 2023 wildfire season. The collected data was used to enhance the sensor and its detection capabilities.
(N5 sensor)
About 450 so-called Alpha and Beta sensors, which can cost thousands of dollars each, have been deployed in areas including Orange County and Bay Area cities to help detect fires in Hawaii, Colorado and Oakland, California. .
The Department of Homeland Security received $4 million in funding over four years to support this effort, but the department has not been able to secure any further federal funding, the agency’s Science and Technology Directorate Sensor Platforms said. said Jeff Booth, director of the technology center.
“We don’t have any more federal funding to take this one step further,” Booth said. “Perhaps the new administration may see the value in rolling this out further.”
A team prepares for a SuperVolo XL flight at Monterey Bay Academy Airport near Watsonville, California.
(Don Ritchie/NASA Ames)
For a startup like Rain, getting buy-in from investors and the fire department is key.
Founded in 2019, Rain operates out of an old control tower at the former Alameda Naval Air Station. The 15-employee company raised $9.7 million in seed financing led by venture capital firm DBL Partners.
Lane hopes to work with Lockheed Martin’s Sikorsky and Orange County fire officials to put the technology into practice.
“Partnerships between innovators and engineers in the firefighting community open up entirely new tools, technologies and markets,” said CEO Brody.
