Wildfires are a significant ecological and economic issue, especially in California, where the total yearly costs can reach several billion dollars (including insurance claims, and rehabilitation and prevention efforts). Distributed multi-agent systems are an attractive solution for both monitoring and containing wildfires for several reasons: robustness to single agent failures, ability to coordinate to adapt to varying conditions, and the advantage of an aerial perspective that can also be used to assist fire-fighters.

The scope of this project includes an appropriate wildfire model and methods for both the control and estimation of a wildfire process.