AI cameras now scanning Minnesota skies for early signs of wildfire
Go Deeper.
Create an account or log in to save stories.
Like this?
Thanks for liking this story! We have added it to a list of your favorite stories.
Artificial intelligence-powered cameras are operating in Minnesota for the first time to detect wildfires before they spread.
Xcel Energy has installed two cameras, located in Mankato and Clear Lake. The cameras are designed to run 24/7 to provide the utility and state agencies real-time information.
Ryan Long, president of Xcel Energy in Minnesota, North Dakota and South Dakota, said this allows for a faster response. Long called the cameras “incredible.”
“They scan a 10-mile radius every minute, and can instantly triangulate the source of smoke,” Long said. The cameras provide alerts to those impacted by the potential wildfires: Xcel Energy, municipal agencies and first responders.
Turn Up Your Support
MPR News helps you turn down the noise and build shared understanding. Turn up your support for this public resource and keep trusted journalism accessible to all.
More camera stations are expected to be installed.
“By the end of 2026, we will have deployed 36 more of those cameras across a diverse geography across our state, in all of our highest wildfire risk zones,” Long said. The cameras capture “what is happening across our system,” he said.
The devices are manufactured by the San Francisco-based Pano AI. Co-founder and Chief Commercial Officer Arvind Satyam said the company has installations in 15 states including Texas, California and Nevada. But the scourge of wildfires is no longer just contained in the western states, Satyam said, pointing to recent wildfires in New Jersey.
Satyam said the cameras can be useful beyond fire season.
“We're seeing hotter hots. Wetter wets. Colder colds. We're seeing a lot of extremes,” he said. “Can we bring technology to be a force multiplier? That was the fundamental question that we asked as a company.”
Satyam said the AI is looking for smoke or the faint whiffs of smoke. He compared it to self-driving cars traveling in a city. They look at objects and roads to figure out if they can continue straight ahead. In this scenario, he said, the cars have a large data set.
“The AI cameras have billions of images,” he said. “We're essentially looking at this landscape continuously to see ‘is that smoke or not smoke?’”
Satyam says the cameras are installed at high vantage points in Mankato and Clear Lake.
“We're looking down on areas of fire risk. We apply AI to detect smoke during the day, heat signatures at nighttime, and then we have human intelligence to validate that,” Satyam said.
