Remote Management Architecture of UAV Fleets for Maintenance, Surveillance, and Security Tasks in Solar Power Plants

Rosende, Sergio Bemposta; Sánchez-Soriano, Javier; Muñoz, Carlos Quiterio Gómez; Andrés, Javier Fernández · 2020 · OpenAlex-citations

DOI: 10.3390/en13215712

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper presents a remote management architecture for fleets of unmanned aerial vehicles (UAVs) designed to automate maintenance, surveillance, and security tasks in solar power plants. The research is motivated by the rapid expansion of global solar capacity and the high costs associated with traditional manual inspections of photovoltaic panels. By leveraging UAVs, the system aims to reduce operational expenses and improve efficiency through automated defect detection, such as identifying hot spots caused by dust or cell damage, as well as enhancing security via pedestrian and vehicle tracking. The proposed system utilizes a distributed, modular architecture comprising UAVs, fixed and onboard cameras, communication networks, and a central processing server. The hardware includes custom multirotor UAVs equipped with PixHawk autopilots, dual-spectrum cameras (RGB and infrared), and onboard processing units like Raspberry Pi Zero W for telemetry and video transmission via 4G/5G networks. The software architecture supports two deployment scenarios: a simple local model for small sites and a multiple architecture for large, distributed plants connected via virtual private networks. Key software modules include navigation, route optimization, and computer vision algorithms for camera calibration, object tracking, and panel inspection. The system integrates both onboard drone cameras and calibrated fixed ground cameras to provide comprehensive environmental monitoring. The study highlights specific technical innovations, particularly the use of convolutional neural networks arranged in series for image segmentation and detection. This approach allows for independent retraining of network layers, facilitating the exchange of higher-resolution cameras without retraining the entire system and ensuring fast response times. The architecture enables the generation of georeferenced thermal maps for defect identification and supports real-time surveillance by differentiating between internal combustion and electric vehicles using thermal thresholds. The system is designed to comply with evolving UAV regulations, utilizing electronic identification and secure communication protocols. The significance of this work lies in its comprehensive integration of hardware and software for autonomous UAV fleet management in renewable energy infrastructure. By automating inspection and security tasks, the architecture offers a scalable solution that reduces reliance on manual labor and specialized equipment. The modular design allows for adaptability across different plant sizes and operational needs, while the use of deep learning and dual-camera systems enhances the accuracy and speed of defect detection and surveillance. This approach contributes to the broader field of smart city and industrial automation by demonstrating how UAV fleets can be effectively managed for critical infrastructure maintenance.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• teleoperation remote driving