Design and Implementation of an Intelligent UAV-Based Border Surveillance System with Autonomous Navigation and Real-Time Monitoring

Abstract

Unmanned aerial vehicles (UAVs) are increasingly adopted for border surveillance because they can deliver persistent coverage, rapid deployment, and real-time situational awareness across wide and remote areas at comparatively low operational cost. This study presents the design, development, and implementation of an intelligent quadcopter-based surveillance platform that combines autonomous navigation with live monitoring and on-board vision processing. The proposed system integrates a Pixhawk 2.4.8 flight controller, GPS waypoint navigation, and FPV video transmission, supported by Python-based computer vision modules for real-time object detection and tracking using OpenCV. To enable end-to-end operation, a custom ground-control web application was developed using React and Java-based REST APIs, with IoT communication protocols to support mission planning, live video viewing, continuous aircraft telemetry/status monitoring, and manual override when required. System validation was performed through a structured set of laboratory and field experiments, including propulsion characterization, ESC and sensor calibration, PID tuning for flight stability, and multi-scenario flight trials. The experimental results indicate stable autonomous flight at altitudes up to 15 m, dependable real-time video streaming, effective detection and tracking of moving targets under practical conditions, and robust fail-safe behavior through return-to-launch during abnormal or emergency events. Overall, the implemented platform provides a scalable and cost-efficient approach to strengthen border-security operations and offers a flexible foundation for future enhancements such as thermal imaging, advanced AI recognition models, and autonomous charging or docking stations.