A Proof-of-Concept Use of Unmanned Aerial Vehicle for Autonomous Person Detection and Tracking in Solar Farms

Abstract

Unmanned Aerial Vehicle (UAV) technology has rapidly advanced, enabling the execution of increasingly complex autonomous tasks. This work presents a proof-of-concept autonomous surveillance framework utilizing a low-cost UAV for indoor monitoring as an initial step toward the surveillance of photovoltaic power plants. The proposed system integrates environment perception, mapping, autonomous navigation, person detection, target tracking, and operator supervision within a unified ROS-based architecture. A lightweight web interface was developed to support mission supervision through autonomous navigation commands, real-time video streaming, and automatic email notifications triggered by detected intruders. Experiments conducted in both simulated and real environments demonstrated the successful integration of the proposed modules, allowing the UAV to map previously unknown environments, autonomously navigate while avoiding obstacles using only monocular visual odometry, and continuously detect and track a target. Despite the limitations imposed by the selected low-cost platform, the results demonstrate the feasibility of the proposed surveillance framework under controlled conditions and establish a modular foundation for future deployment in photovoltaic power plants through the adoption of industrial UAV platforms, long-range communication infrastructures, and multi-sensor localization techniques.