Universities are developing autonomous smoke-sensing drones to study fire behavior by analyzing smoke patterns, a move toward better wildfire prediction and management.

The drones currently work in smaller-scale tests with about 25 minutes of flight time and performance dips in colder temperatures, signaling hurdles before broad deployment.

In a recent field trial at Cedar Creek, a swarm of autonomous drones carrying upgraded sensors and propulsion gathered particle data and mapped smoke flow.

This work sits within a broader shift toward AI-assisted fire detection and management, alongside NOAA’s Next-Generation Fire System and AI-enabled cameras by utility partners, highlighting growing use of autonomous tools in fire science.

Funding comes from the University of Minnesota and the National Science Foundation, underscoring institutional and federal support for autonomous environmental monitoring.

The system uses one central drone coordinating four others to navigate smoke and collect data from multiple elevations, advancing understanding of particle transport and vertical smoke distribution.

The research aims to predict fire movement, guide firefighting strategies, and improve surveillance during wildfires and prescribed burns, with future goals including longer battery life and more data collection.

Experts note the gap between controlled lab work and messy real-field conditions, but autonomous drone systems offer a path to comprehensive surveillance and a deeper grasp of smoke transport and fire behavior.