Autonomous Aerial System for Solar Panel Maintenance
Keywords:
Artificial Intelligence (AI), Convolutional Neural Networks (CNN), Solar energy, Solar panels, Sustainable energyAbstract
This study presents an innovative solution for the automated maintenance of solar panels using an autonomous aerial drone system. The paper integrates Edge AI and Convolutional Neural Networks (CNN) for real-time dust detection and cleaning of solar panels. The drone captures images of solar panels using an onboard camera, processes the images using a pre-trained CNN model on an edge device, detects dust accumulation, and triggers a water-spraying mechanism to clean the panels. This system enhances the efficiency of solar power generation by maintaining panel cleanliness, especially in remote or hard-to-reach locations. The proposed system integrates a quadcopter equipped with a high-resolution camera for real-time dust detection using AI-based image processing. The quadcopter utilizes machine learning algorithms to assess dust levels and initiate cleaning processes accordingly. It features a precise water spraying mechanism, controlled via a web-based dashboard, allowing both autonomous and manual operation. The system’s cloud-based analytics platform provides real-time monitoring, data logging, and optimization of cleaning schedules based on environmental factors, including weather and panel usage.
References
M. Maghami, H. Hizam, and C. Gomes, “Impact of dust on solar energy generation based on actual performance,” 2014 IEEE International Conference on Power and Energy (PECon), Kuching, Malaysia, 2014, pp. 388-393, doi: https://doi.org/10.1109/PECON.2014.7062476
Y. T. Zhang, J. T. Zhang, Y. Liu, K. Liu, and Q. Wang, “Spray-on steady-state study of multi-rotor cleaning unmanned aerial vehicle in operation of photovoltaic power station,” Energy Reports, vol. 11, pp. 5638–5653, Jun. 2024, doi: https://doi.org/10.1016/j.egyr.2024.04.060
N. M. Haegel and S. R. Kurtz, “Global progress toward renewable electricity: Tracking the role of solar (Version 4)," in IEEE Journal of Photovoltaics, vol. 15, no. 2, pp. 206-214, March 2025, doi: https://doi.org/10.1109/JPHOTOV.2024.3450020
S. Rehman, M. A. Mohandes, A. E. Hussein, L. M. Alhems, and A. Al-Shaikhi, “Cleaning of photovoltaic panels utilizing the downward thrust of a drone,” Energies, vol. 15, no. 21, p. 8159, Nov. 2022, doi: https://doi.org/10.3390/en15218159
T. Nasim, “The impact of electricity price on residential solar PV adoption: A longitudinal analysis of economic and social drivers,” Openarchive.usn.no, 2024, doi: no. usn:wiseflow:7236733:61127697
M. Al-Housani, Y. Bicer, and M. Koç, “Experimental investigations on PV cleaning of large-scale solar power plants in desert climates: Comparison of cleaning techniques for drone retrofitting,” Energy Conversion and Management, vol. 185, pp. 800–815, Apr. 2019, doi: https://doi.org/10.1016/j.enconman.2019.01.058
Y. T. Zhang, J. T. Zhang, Y. Liu, K. Liu, and Q. Wang, “Spray-on steady-state study of multi-rotor cleaning unmanned aerial vehicle in operation of photovoltaic power station,” Energy Reports, vol. 11, pp. 5638–5653, Jun. 2024, doi: https://doi.org/10.1016/j.egyr.2024.04.060
K. Mahroof, A. Omar, N. P. Rana, U. Sivarajah, and V. Weerakkody, “Drone as a service (DaaS) in promoting cleaner agricultural production and circular economy for ethical sustainable supply chain development,” Journal of Cleaner Production, vol. 287, p. 125522, Mar. 2021, doi: https://doi.org/10.1016/j.jclepro.2020.125522
D. Feldman, K. Wu, and R. Margolis, “H1 2021 solar industry update,” OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information), Jun. 2021, doi: https://doi.org/10.2172/1808491
A. Ashraf and Muneeb Sagheer, “Renewable energy capacity and technological innovations: A review of global trends and future directions,” Environmental Progress & Sustainable Energy, Jul. 2025, doi: https://doi.org/10.1002/ep.70071
