Drone Technology Inspects Solar Panel with AI-Powered Fault Detection

Post-event drone surveys quantify hail, wind, or flood damage within hours, guiding rapid response and repair bids.

Norwegian researchers demonstrate how UAV swarms and machine learning can speed maintenance of massive solar installations

As utility-scale solar power plants continue their rapid expansion worldwide, a critical challenge has emerged: how to efficiently inspect millions of photovoltaic modules spread across vast installations. A breakthrough study by Norwegian researchers has revealed that combining unmanned aerial vehicle (UAV) swarms with advanced machine learning algorithms could revolutionize solar panel maintenance, potentially saving the industry billions in operational costs.

Published in the IEEE Journal of Photovoltaics in September 2025, the research by V. Lofstad-Lie and colleagues at leading Norwegian institutions demonstrates how the latest YOLO11 artificial intelligence model can detect solar panel defects with remarkable accuracy from altitudes up to 80 meters—significantly higher than previously possible.

The Challenge of Scale

The scale of modern solar installations has grown exponentially. Utility-scale solar power plants can now have more than a million PV modules, creating an unprecedented maintenance challenge. Traditional manual inspection methods, which involve workers walking through installations with handheld thermal cameras, are becoming increasingly impractical for these massive facilities.

Conventional manual inspections require temporary shutdown of the PV system that causes lost revenue, and putting off an inspection for too long may result in significantly decreased energy production due to undetected faulty equipment. In contrast, drone solar inspections are reported to be 40% cheaper than manual solar inspections, and critically, can be performed while the system remains connected to the grid.

AI Meets Aerial Inspection

The Norwegian team's key innovation lies in training the YOLO11 machine learning model—the latest iteration of the "You Only Look Once" object detection algorithm—specifically for thermal image analysis of solar panels. This is the first time this architecture has been trained and tested using thermal and optical images of solar panels, representing a significant technological advancement.

Recent comparative studies have shown impressive performance across different YOLO versions for solar fault detection. YOLOv11 delivered the highest mAP@0.5 (93.4%), demonstrating balanced performance across defect categories, while YOLOv5 achieved the fastest inference time (7.1ms per image) and high precision (94.1%) for cracked panels.

The 80-Meter Sweet Spot

One of the study's most significant findings addresses a critical tradeoff in aerial inspections: the balance between efficiency and accuracy. The researchers discovered that fault detection remained robust up to approximately 80 m, but georeferencing error became the primary limiting factor at higher altitudes.

This 80-meter threshold represents a breakthrough for the industry. Current best practices typically require much lower flight altitudes to achieve the necessary image resolution. For a detailed IEC inspection, the drone should fly at a low enough altitude to achieve roughly a 3 cm per pixel ground sampling distance, often requiring flying relatively close to the panels.

The Norwegian research demonstrates that advanced AI can maintain detection accuracy at higher altitudes, potentially transforming inspection efficiency. Higher flight altitudes mean larger coverage areas per flight, reduced battery consumption, and the ability to cover vast solar installations more quickly.

Swarm Intelligence Takes Flight

Perhaps the most futuristic aspect of the research involves UAV swarm technology. Swarm technology allows more than one drone to simultaneously conduct large-scale solar farm inspections, significantly bringing down the time of inspection without compromising the quality of data.

The concept of autonomous UAV swarms for solar inspection represents a paradigm shift. A multiprotocol architecture is developed to ensure effective communication and coordination between IoT Modules and UAVs, facilitating seamless data exchange and system integration. This integrated approach combines on-panel sensors with aerial inspection capabilities, creating a comprehensive monitoring ecosystem.

Two-Stage Strategy Cuts Costs

Complementing the Norwegian research, other recent studies have developed optimized inspection strategies that could reduce operational costs by up to 70%. A two-stage autonomous flight strategy is adopted where the first stage images the entire park quickly at low resolution, flying fast at high altitude to detect possible faults. In the second stage, the detections are imaged at higher resolution, by flying an optimized path at lower altitude, for fault classification.

This approach represents a fundamental shift from traditional comprehensive scanning to intelligent, targeted inspection. Based on actual data, including a park-scale survey, this strategy is shown to have potential for significant savings in operation time, on the order of 70% or more, at realistic fault densities.

Industry Standards Drive Adoption

The rapid advancement of drone inspection technology is being supported by evolving industry standards. The inspection must be up to IEC TS 62446-3 thermography standards set by the International Electrotechnical Commission, which provides a framework for ensuring consistent, reliable results across the industry.

Aerial thermography provides a 95-99% accuracy rate in the detection of PV system anomalies and defects affecting performance, making it increasingly attractive to solar plant operators and investors seeking to maximize returns on their renewable energy investments.

Real-World Performance Metrics

The practical impact of these technological advances is already being demonstrated in commercial deployments. Modern types of UAVs can detect nearly 95% of the common solar panel issues like physical damage, electrical faults, and hot spots with the use of advanced thermal imaging technology.

Speed improvements are equally impressive. Typical aerial inspection drones can cover up to 10 MW per hour, whereas manual inspections would be lucky to cover 1 MW in 10 hours of work, representing a dramatic increase in inspection efficiency.

Advanced Sensors and Future Developments

The technology continues to evolve rapidly. Advanced sensors will continue to trend, including LIDAR and hyperspectral imaging, offering enhanced insights into the structural integrity and panel performance. These developments promise even more detailed analysis capabilities, potentially detecting issues before they become visible in thermal imaging.

Predictive maintenance, powered by historical drone inspection data, will reduce both maintenance costs and downtime, suggesting that the future of solar maintenance will be increasingly proactive rather than reactive.

Global Implementation Challenges

Despite the promising technology, implementation faces several challenges. Australian UAV regulations mandate Remote Pilot License certification and commercial operations approval, and similar regulatory frameworks are developing worldwide. Different countries have varying requirements for drone operations, particularly over critical infrastructure like solar installations.

Technical challenges also remain. Despite strong performance for common defects like dusty panels (mAP@0.5 > 98%), bird drop detection posed challenges due to dataset imbalances, highlighting the need for more comprehensive training datasets for AI systems.

Economic Impact and Future Outlook

The economic implications of these technological advances are substantial. The O&M annually costs of a traditional inspection are estimated to be around 11.30 €/kWp, from which only 1.50 €/kWp are the costs associated to the planned monitoring and inspection. The dramatic efficiency improvements offered by AI-powered drone inspections could significantly reduce these costs.

As the solar industry continues its exponential growth, with installations reaching unprecedented scales, the Norwegian research and related developments represent a critical technological evolution. The combination of high-altitude inspection capabilities, AI-powered fault detection, and swarm technology could enable the maintenance of solar installations at scales previously thought impossible.

The research demonstrates that the future of solar maintenance lies not just in automation, but in intelligent automation that can adapt to the unique challenges of utility-scale renewable energy infrastructure. As these technologies mature and regulatory frameworks adapt, they promise to make solar energy even more cost-effective and reliable, supporting the global transition to renewable energy.

---

Sources

1. Lofstad-Lie, V., Simonsen, A., Nygaard, T. F., & Marstein, E. S. (2025). Data Quality Analyses for Automatic Aerial Thermography Inspection of PV Power Plants. IEEE Journal of Photovoltaics, 15(5), 722-731. doi: 10.1109/JPHOTOV.2025.3587297. Retrieved from https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11091333&isnumber=11133592
2. "6 Major Benefits of Aerial Thermography for Solar Inspections." The Drone Life. Retrieved from https://thedronelifenj.com/benefits-of-infrared-drone-solar-inspections/
3. Buerhop-Lutz, C., et al. (2023). Infrared thermography monitoring of solar photovoltaic systems: A comparison between UAV and aircraft remote sensing platforms. ScienceDirect. Retrieved from https://www.sciencedirect.com/science/article/pii/S2451904923007321
4. "Cost-Effective Flight Strategy for Aerial Thermography Inspection of Photovoltaic Power Plants." IEEE Journals & Magazine. Retrieved from https://ieeexplore.ieee.org/document/9877756/
5. "Cost-effective flight strategy for aerial thermography inspection of photovoltaic power plants." IEEE Conference Publication. Retrieved from https://ieeexplore.ieee.com/document/9518710/
6. "The Ultimate Guide to Aerial Thermographic Inspections for Solar Plants." Above Surveying. (2024, April 22). Retrieved from https://www.abovesurveying.com/an-introduction-to-aerial-thermographic-inspection/
7. "How to Inspect Solar PV Systems with Drones (UAVs) Part 2." Raptor Maps. Retrieved from https://raptormaps.com/blog-posts/how-to-inspect-solar-pv-systems-with-drones-(uavs)-part-2
8. "A Beginner's Guide to Aerial Solar Inspections." 60 Hertz Energy. (2024, September 27). Retrieved from https://60hertzenergy.com/how-to-aerial-solar-inspection/
9. Khanam, R., & Hussain, M. (2025). Detecting Defects in Solar Panels Using the YOLO v10 and v11 Algorithms. Electronics, 14(2), 344. Retrieved from https://www.mdpi.com/2079-9292/14/2/344
10. Khanam, R., Asghar, T., & Hussain, M. (2025). Comparative Performance Evaluation of YOLOv5, YOLOv8, and YOLOv11 for Solar Panel Defect Detection. Preprints. Retrieved from https://www.preprints.org/manuscript/202501.0788/v1
11. "Real-time PV Fault Detection using Embedded Machine Learning." IEEE Conference Publication. Retrieved from https://ieeexplore.ieee.org/document/10640018
12. Khanam, R., Asghar, T., & Hussain, M. (2025). Comparative Performance Evaluation of YOLOv5, YOLOv8, and YOLOv11 for Solar Panel Defect Detection. MDPI. Retrieved from https://www.mdpi.com/2673-9941/5/1/6
13. "SPHERE: Benchmarking YOLO vs. CNN on a Novel Dataset for High-Accuracy Solar Panel Defect Detection in Renewable Energy Systems." MDPI. (2025, April 28). Retrieved from https://www.mdpi.com/2076-3417/15/9/4880
14. Hussain, M., et al. (2023). Fault detection from PV images using hybrid deep learning model. ScienceDirect. Retrieved from https://www.sciencedirect.com/science/article/abs/pii/S0038092X23008411
15. Nassreddine. (2025). Fault Detection and Classification for Photovoltaic Panel System Using Machine Learning Techniques. Wiley Online Library. Retrieved from https://onlinelibrary.wiley.com/doi/pdf/10.1002/ail2.115
16. "An approach based on deep learning methods to detect the condition of solar panels in solar power plants." ScienceDirect. (2024, March 2). Retrieved from https://www.sciencedirect.com/science/article/abs/pii/S0045790624000715
17. "Fault diagnosis of photovoltaic systems using artificial intelligence: A bibliometric approach." ScienceDirect. (2023, October 26). Retrieved from https://www.sciencedirect.com/science/article/pii/S2405844023086991
18. "Key Insights for Drone Solar Panel Inspection in 2025." AUAV. (2025, June 30). Retrieved from https://www.auav.com.au/news/key-insights-for-drone-solar-panel-inspection-in-2025/
19. "The Best Drone Solar Inspection Software for 2025." The Drone Life. Retrieved from https://thedronelifenj.com/drone-solar-inspection-software/
20. Lagrange, A., et al. (2018). On the technologies empowering drones for intelligent monitoring of solar photovoltaic power plants. ScienceDirect. Retrieved from https://www.sciencedirect.com/science/article/pii/S1877050918310366
21. Delponte, E., et al. (2025). Framework for autonomous inspection of PV plants using IoT electronics on each PV panel and UAV collaboration. ScienceDirect. Retrieved from https://www.sciencedirect.com/science/article/pii/S2590174525000108
22. "How to Inspect Solar PV Systems with Drones (UAVs) Part 1." Raptor Maps. Retrieved from https://raptormaps.com/blog-posts/how-to-inspect-solar-pv-systems-with-drones-(uavs)-part-1
23. Aghaei, M., et al. "A comprehensive review of unmanned aerial vehicle-based approaches to support photovoltaic plant diagnosis." PMC. Retrieved from https://pmc.ncbi.nlm.nih.gov/articles/PMC10789596/
24. Tsanakas, J. A., et al. (2023). Solar Photovoltaic Hotspot Inspection Using Unmanned Aerial Vehicle Thermal Images at a Solar Field in South India. MDPI. Retrieved from https://www.mdpi.com/2072-4292/15/7/1914
25. "Complete Drone Solar Panel Inspection Guide 2025." Averroes AI. Retrieved from https://averroes.ai/blog/drone-solar-panel-inspection-guide
26. Koester, L., et al. (2025). Review of unmanned ground vehicles for PV plant inspection. ScienceDirect. Retrieved from https://www.sciencedirect.com/science/article/pii/S0038092X25001677
27. "A UAV infrared measurement approach for defect detection in photovoltaic plants." IEEE Conference Publication. Retrieved from https://ieeexplore.ieee.com/document/7999594/
28. Liu, Y., et al. (2022). Accuracy Assessment of a UAV Direct Georeferencing Method and Impact of the Configuration of Ground Control Points. MDPI. Retrieved from https://www.mdpi.com/2504-446X/6/2/30
29. "Solar Panel Inspection Drones." Advexure. Retrieved from https://advexure.com/pages/solar-panel-inspection-drones
30. Data Quality Analyses for Automatic Aerial Thermography Inspection of PV Power Plants | IEEE Journals & Magazine | IEEE Xplore