The fire detection devices come from California- and Florida-based Illumination Technologies, which launched in 2019 using cameras that had been designed for the German space program, according to Illumination Technologies California CEO Chris Eldridge.
He said the technology began proving itself before the recent agreement with Napa County, which may be most famous for its wine but is perhaps becoming increasingly known for its wildfires. In 2020, in fact, wildfires burned some 42 percent of that county, according to one estimate. That’s more than 200,000 acres.
“We detected a significant fire last year, a fire that occurred at night,” Eldridge said of his company’s technology, which can spot fires not only via optical data but through sensory information tied to heat and gas. “We located the fire down to a few yards.”
The deployment reflects the growing market for disaster relief efforts tied to AI and machine learning — a market that has gained more of the spotlight since the pandemic is attracting not only younger companies but giants such as Amazon.
MARKET TRENDS
For instance, Napa County has decided to deploy three cameras from Illumination, the first major contract for the subscription-fee operation. Napa reportedly is paying $6,000 per camera per month. The cameras are hardly the only line of defense, of course, for that government agency. The cameras help county officials better plan their wildfire response plans and respond more precisely to specific fires.
But look beyond the present and that relatively small deployment and you’ll see optimistic signs for the future of AI-powered disaster management and relief technology — and why companies such as Illumination are bullish about the prospects of their own offerings.
In the U.S., sales for crisis, emergency and incident management platforms will grow by a nearly 9 percent compound annual growth rate over the next decade, according to Persistence Market Research. AI promises to play a significant role in that growth.
“AI is being used to analyze past data to predict what is likely to happen in the event of a disaster, and that data can be integrated with online dashboards so that emergency personnel can respond in real time,” according to that firm. “Also, data analysis and machine learning can be used to identify locations affected by quakes that have not yet been assessed or received assistance.”
That predication of growth comes amid ongoing calls for more federal funding of research into AI disaster-related technology, and a growing number of global case studies about how AI can offer significant monitoring and relief tasks. Those tools could prove especially useful as countries and smaller units of government prepare to deal with the impacts of climate change, according to those reports.
“AI allows responders to monitor the disaster in real-time. Whether it’s doctors responding to Ebola outbreaks or firefighters rushing to put out an enormous flame, AI makes it possible to view the situation unfolding in real-time,” reads part of a blog posting from Nix Solutions. “The responders can also get up-to-date alerts about what’s happening. Processing this information allows the decision-makers to determine the best course of action depending on the disaster’s location and severity, among other factors.”
DIFFERENT PLAYERS
A quick look at the market shows the variety of business possibilities for AI disaster management and relief technology for public agencies.
California-based software provider One Concern, for example, recently teamed with California’s CoreLogic on a project using AI for disaster mitigation efforts. CoreLogic sells property data and analytics to governments, real estate professionals, banks and others. That data includes information about floods, winds, storm surges and climate change.
The idea is to combine the CoreLogic data with One Concern’s “AI-enabled resilience solutions and disaster-risk reduction technologies to address and predict weather hazards and escalating climate threats amidst an increasing global focus to develop environmental, social, governance and resilience goals,” reads a recent statement from One Concern.
Amazon, too, has a presence in this general area, if one takes into account machine learning.
According to the Amazon Web Services Machine Learning Blog, the company has worked with university researchers in the U.S. to better analyze disaster scene imagery in hopes of improving disaster relief and damage assessment efforts.
“It’s easy to distinguish a lake from a flood. But when you’re looking at an aerial photograph, factors like angle, altitude, cloud cover and context can make the task more difficult,” wrote Morgan Dutton, a senior program manager for Amazon, in that blog. “And when you need to identify 100,000 aerial images in order to give first responders the information they need to accelerate disaster response efforts? That’s when you need to combine the speed and accuracy of machine learning with the precision of human judgment.”
VITAL FOR FIRES
Meanwhile, groups involved in public safety are starting to tout the need to use AI to anticipate and manage disasters — another point in favor of market growth and the prospects of participating companies.
A recent example comes from the Western Fire Chiefs Association (WFCA).
“You could not hire enough firefighters,” stated Brent VanKeulen, deputy director of the WFCA. “You can’t fly enough planes. You can’t get enough dozers on the ground to meet the challenge of what we’re facing now.”
That’s a motivation guiding Illumination, whose cameras were originally used to spot such astronomical features as comet gas, according to Eldridge, the CEO.
As he tells it, the AI component is vital going forward to better protect against wildfires.
“Without it, it would basically be just overloaded human monitoring,” he said. “AI is what makes it a proactive system that allows for efficient monitoring. AI is what allows limited resources to be pointed the most efficient way.”
