WVU engineers including Hang Woon Lee, left, and Brycen Pearl have developed a satellite positioning system that improves the detection of wildfires from space.
Photo Credit: WVU Photo/Brian Persinger
Scientific Frontline: Extended "At a Glance" Summary: WildFIRE-DS AI Satellite System
The Core Concept: WildFIRE-DS (WildFire-applicable Intelligent and Responsive Ensemble for Detection and Scheduling) is an artificial intelligence framework designed to enable satellite constellations to autonomously interpret wildfire imagery and dynamically adjust their positions for continuous, near-real-time monitoring.
Key Distinction/Mechanism: Unlike standard satellite networks restricted to static observation schedules, this AI framework uses interpreted imagery and statistical models to automatically retask and coordinate a cooperative group of satellites, ensuring they rapidly revisit and track fast-spreading fires.
Major Frameworks/Components:
- AI-Driven Image Interpretation: Processes and validates the existence of wildfires autonomously directly on the satellite.
- Ensemble Scheduling Algorithm: Coordinates large groups of satellites to share information and track complex environmental targets collaboratively.
- Autonomous Retasking: Permits satellites to reposition and deviate from initial deployment routes to optimize viewing angles over newly detected hotspots.
Branch of Science: Aerospace Engineering, Artificial Intelligence, Remote Sensing, and Environmental Science.
Future Application: Facilitating the deployment of high-resolution satellite constellations that can automatically dispatch firefighters upon detecting a blaze, effectively eliminating the reliance on civilian emergency 911 reports.
Why It Matters: Wildfires can spread at speeds up to 20 mph and create their own localized weather systems; autonomous tracking significantly accelerates emergency response, which is critical for saving lives and minimizing catastrophic infrastructure damage.
A new artificial intelligence system developed by West Virginia University engineers could help firefighters respond to wildfires sooner by enabling satellites to detect blazes and automatically adjust their positions for continued monitoring.
Unlike drones and ground-based sensors, satellites can monitor vast areas of the planet without requiring local infrastructure or routine maintenance. WVU researchers Brycen Pearl, Joshua Warner, and Hang Woon Lee developed a framework that allows satellites to not only detect wildfires but also coordinate with one another and adjust their observation schedules as fires spread.
“Wildfires move quickly—as fast as 15 to 20 mph under the right conditions—and major wildfires can cover hundreds of thousands of acres,” according to Lee, director of the WVU Space Systems Operations Research Laboratory and assistant professor at the WVU Benjamin M. Statler College of Engineering and Mineral Resources.
“Both of these aspects make containment very difficult, and the rapid rate at which wildfires change makes them very hard to track as well. So does the terrain, since wildfires thrive in areas with dense vegetation and hills. Other researchers are improving the response in many ways, including ground networks and drones, but those are limited to the areas where they are deployed.”
Satellites, on the other hand, can see the entire planet in a few days or less. And even better, Lee said, is a cooperative group of satellites sharing information and passing over hotspots.
In addition to gathering detailed imagery of fire conditions on Earth, satellites also carry sensors that collect information on vegetation, surface temperatures, wind patterns, and a myriad of conditions that make wildfires so complicated and hard to predict.
“Wildfire behavior is a complex system in which a huge number of factors interact with each other in ways that can spiral into unexpected outcomes,” said Pearl, a Statler College doctoral candidate in aerospace engineering.
“Wind is the biggest driver of how fires spread, and it’s notoriously unpredictable. For instance, a fire burning in a canyon can generate its own wind through a chimney effect, pulling air in from below and blasting flames up and out at great speeds. Also, large fires get so hot that they can change the atmosphere above them, creating clouds that are the fire’s own personal thunderstorm.”
That is why the WVU team developed an AI framework for interpreting satellite images of wildfires, ensuring the accuracy of the interpretations with the help of statistics, and then automatically retasking and repositioning satellites for continued monitoring.
Pearl led the development of the WildFire-applicable Intelligent and Responsive Ensemble for Detection and Scheduling, or WildFIRE-DS, with Lee and undergraduate researcher Warner. The algorithm they developed with support from the NASA West Virginia Established Program to Stimulate Competitive Research is presented in a paper in the Journal of Aerospace Information Systems.
Warner explained that because a tiny spark can grow to consume hundreds of acres in under an hour, wildfire detection systems must be just as fast.
“Satellites need to pass overhead frequently, ground sensors need to be in constant operation, and data interpretation needs to occur in near real time,” he said.
Warner pointed to the 2025 Palisades Fire, which burned 23,448 acres in California, claimed twelve lives, destroyed 6,837 structures, and caused more than $25 billion in damages. But losses would have been even worse, he said, if first responders had not been able to save precious time thanks to innovations like the ALERTCalifornia AI Camera Network, which links more than 1,200 high-definition cameras with near-infrared vision to provide 24-hour backcountry monitoring.
Take that concept to the next level, Pearl added, and the network of cameras becomes a constellation of satellites—“big groups of satellites flying in formation in space, like Earth Fire Alliance’s FireSat and the OroraTech Wildfire Constellation.”
“These constellations are planned to have fifty to one hundred satellites with the resolution to see fires as small as cars, powered by AI to interpret many previous images of the same area to ensure the existence of a wildfire before automatically sending out firefighters—no need for someone to call 911 first,” he said.
While the FireSat and OroraTech Wildfire satellite constellations will use AI to interpret satellite imagery and validate those interpretations, the WVU WildFIRE-DS framework adds additional capacity: using the interpreted imagery to autonomously set a new schedule for satellite positioning and monitoring.
That means the satellites can be repositioned, ensuring they are in the best position possible to see newly detected wildfires, rather than being forced to stay where they were initially deployed. By allowing satellites to move, WildFIRE-DS lets them revisit a desired location quickly and see the wildfires more frequently.
“On the ground, teams are deploying permanent sensor and camera systems that watch fire-prone land at all times,” Pearl said. “In the air, drone technology is reaching new heights with better drone propellers, smaller frames, better battery life, better cameras, and more well-trained remote pilots. In space, satellites are being dedicated to wildfire monitoring with better placement, better cameras, and AI to process images and detect wildfires on the satellite itself before relaying that back to the ground.”
All the innovations share a common goal, he added—“giving firefighting crews a head start.”
Published in journal: Journal of Aerospace Information Systems
Authors: Brycen D. Pearl, Joshua G. Warner and Hang Woon Lee
Source/Credit: West Virginia University
Edited by: Scientific Frontline
Reference Number: ai071326_01