As wildfires grow larger, faster and impact more communities across Canada, researchers at the University of Toronto Institute for Aerospace Studies (UTIAS) are developing new technology that could give first responders a critical advantage: the ability to predict how a fire will evolve hours into the future.
Professor Steven Waslander (UTIAS) is working with researchers at Natural Resources Canada and Simon Fraser University on TankerVision — a project aimed at creating more accurate, AI-powered fire prediction models.
Such tools could help inform key decisions about firefighting resource allocation and evacuations.
The team has partnered with government agencies and industry, including BC Wildfire, Alberta Wildfire, Canadian Forestry Services, aircraft operator Conair Aerial Firefighting, and Voxelis, a wildfire monitoring service provider.
“Wildfire growth is currently estimated using 15 to 20 prescribed burns from the 1980s, plus data on weather, wind and forest conditions, all fused with decades of human experience,” says Waslander.
“But this method hasn’t been adapted to current fire severity increases due to climate change and involved a great deal of extrapolation from a limited dataset.”
“Our big idea here is to use AI and computer vision to capture unprecedented numbers of fires from the air to build more sophisticated, data-driven wildfire prediction models.”
To do this, high‑resolution colour and infrared cameras and computers are installed inside the piloted aircraft that are currently used to monitor and fight fires.
In 2024, the team began collecting data onboard a bird-dog aircraft, which circles above wildfires to guide water bombers. These planes spend hours tracking the fire line, making them ideal for gathering continuous, high‑quality footage.
“Once the plane is airborne and reaches a certain speed, it begins scanning for smoke or flames and records only when a fire is visible,” says Waslander.
“All the storage is on board the aircraft so our partners can review the data before we start working on it to make sure we aren’t logging any confidential information.”
First, each image is segmented by AI to identify which pixels correspond to fire, smoke or ground. This is trickier than it sounds, as capturing the extent of smoke in an image can be difficult for modern segmentation techniques.
The task poses similar challenges to autonomous navigation in adverse weather conditions — variables Waslander is already familiar with as the lead investigator of the WinTOR all-weather driving program.
“The methods we use are all related to autonomous driving techniques — it’s the same labeling, object detection and segmentation tasks, but this time the challenge is to find the fire boundary,” says Waslander.
“Here, we use segmentation models trained on all kinds of other challenges for segmentation, and then we fine-tune and adapt them to the small amount of data we have available to us. We’re collecting a lot of data from the fire perspective, but from the AI perspective, it’s still a tiny data set.”
Once segmented, the system maps the images onto 3D terrain by aligning the image recordings with satellite maps. This step relies on expertise from field robotics, where aligning images with older or imperfect maps is a familiar challenge. The merged data reveals how the fire is moving over time.
The team will then combine these measurements with weather data and wind readings to build AI models capable of forecasting fire growth.
“Last season, we logged 50 fires, which is currently the highest number recorded in one season,” says Waslander.
“This year, we plan to expand the data collection efforts to three aircraft and hope to capture more than 100 wildfires, sufficient to start developing and validating our prediction models.”
The team is now preparing hardware for this fire season and hopes to expand partnerships across the country, including in Ontario, to help transform how Canada responds to wildfires.
The team’s first research paper, focused on segmentation, has been submitted, with a full-system paper and public dataset planned for release in fall 2026.
Waslander says the project resonates deeply with students and researchers who have witnessed the increase in severe fires firsthand.
“I think everyone involved in the project really feels passionately about the direction we’re going and see it as an opportunity to make a real impact on keeping us safe in the years to come as things get hotter,” says Waslander.
“The contributions to Canadian safety and preservation of our country are really motivating and it’s nice to be able to think our research can actually contribute to this.”
