In 2023, more than 7,000 wildfires across California burned more than 330,000 acres, took four lives, and destroyed more than 150 structures. Loyola Marymount University Professor Gustavo Vejarano has embarked on a groundbreaking project that could transform combating wildfires.
Thanks to a grant of nearly $200,000 over two years from the National Science Foundation, Vejarano’s project, titled “ERI: Fault-Tolerant Monitoring of Moving Clusters of Targets using Collaborative Unmanned Aerial Vehicles,” aims to harness the power of collaborative drones to monitor ground activities, particularly during wildfires, with unprecedented efficiency.
“The wildfires in California are a pressing issue, and current methods are heavily reliant on human calculations,” Vejarano emphasized.
Vejarano’s project addresses a significant challenge: ensuring that small drones can operate continuously and troubleshoot, effectively collaborating. His vision? A fleet of drones working in harmony to monitor situations on the ground, sharing critical information seamlessly, and adapting to challenges autonomously.
“By making sure these drones can fly for long periods without central command, and even if things do happen, to help them adapt to those changes, will ensure they work together effectively,” Vejarano said.
Now, one year into this ambitious project, Vejarano has managed to move the needle on discovering the formula to reach his goals. With the help of eight students, progress has been slow but steady. The grant provided funding for hardware and software components of the project, creating a unique learning opportunity for students in electrical engineering, computer engineering, and computer science.
“Students are actively involved in this real-world problem-solving endeavor,” said Vejarano. “From data acquisition to flight control, they are breaking down the challenges into manageable tasks, gaining invaluable experience in the process.”
Much like the drones, the student researchers are intended to operate independently and as a group. Three of those students were graduate research assistants who managed teams of undergraduate researchers — an exceptional, hands-on experience for students earning their bachelor’s degree.
One student, Alex Alvarez ’25, said he appreciated the interdisciplinary work he gets to do with other Seaver College students. “I’m a computer science major, but many in my team are engineering students. Learning about their work and how to integrate your own broadens your understanding of how your interests apply to larger projects. Learning about AI and machine learning has been fun but getting to apply that knowledge to a real-world problem has proved to be rewarding.”
One of the project’s challenges is developing a sophisticated drone system capable of autonomous flight and real-time data collection. The drones, equipped with onboard computers and GPS (as Vejerano put it “a laptop that flies”), can navigate using waypoints and capture thermal images. One graduate capstone project successfully integrated a thermal camera with a drone, enabling it to collect telemetry data, including coordinates, altitude, and temperature, to identify hotspots on the ground.
The project leverages complex mathematics and technology, including game theory and Gaussian Mixture Models, to enable collaboration among drones. By using game theory, the drones make autonomous decisions to maximize their “mission value” without central command, enhancing the system’s resilience when there is more than one possible point of failure. Vejarano’s team focuses on optimizing flight paths to minimize energy consumption and time, addressing challenges such as limited flight times and harsh environmental conditions.
In collaboration with the U.S. Forest Service, Vejarano’s research will eventually be tested in real-world conditions, monitoring prescribed fires and controlled burns in San Dimas Experimental Forest. The drones, acting as a cohesive unit, will detect and monitor the spread of fires, revolutionizing our approach to wildfire management. So far, successful tests using hot water have been performed in designated drone flight areas near LMU.
The project’s potential applications extend beyond wildfire monitoring. Vejarano envisions an autonomous network capable of addressing other scenarios that threaten the safety of first responders and disaster management personnel.
