Integrated UAV Platform for Multi-Spectral, Thermal, and EOS Imaging in Wildfire Monitoring and Modeling

This thesis presents the design and development of a modular unmanned aerial vehicle (UAV) system based on a quadcopter platform for flexible and efficient wildfire-related multimodal image acquisition. Addressing key limitations in ecological UAV monitoring such as sensor inflexibility, time-consuming reconfiguration, and imprecise image georeferencing, the system introduces a versatile payload integration framework supporting three distinct imaging sensors: MicaSense Altum-PT, FLIR Vue Pro R, and Sony Alpha 6000.All onboard components, including the flight controller, autopilot software, GNSS module, motors and ESCs, were selected to optimize stability and payload performance. A gimbal-free, downward-facing mount simplifies field deployment, while custom integration enables precise geotagging. Field tests were conducted over grassland environment using mission-based corridor scans. The resulting imagery was processed in PIX4Dmapper to generate dense point clouds, orthomosaics, digital surface models (DSMs), and vegetation indices. A comprehensive evaluation of flight performance and data quality was performed to validate the system’s effectiveness. The proposed platform enhances UAV-based ecological monitoring in pre- and post-wildfire scenarios by supporting rapid sensor swapping and robust data collection. It facilitates high-resolution analysis of vegetation, soil, and burn impact, contributing to improved wildfire assessment and post-fire ecosystem recovery monitoring. This research contributes to the National Science Foundation (NSF) EPSCoR project “Harnessing the Data Revolution for Fire Science (HDRFS)” through the Cyberinfrastructure Innovations (CII) component.