Abstract
The rapid growth of the drone market presents both opportunities and challenges, particularly in the area of energy management. One of the main limitations in drone operation is the restricted flying range, necessitating frequent battery recharges. A scalable and efficient solution for persistent operation is the establishment of a network of wireless recharging stations, particularly located on building rooftops. Resonant wireless power transfer (WPT) is an ideal technology for this purpose, enabling non-contact energy transfer. However, the variability in the coupling factor between thedrone and the recharging pad due to misalignments poses a significant technical challenge, leading to potential inefficiencies in power transfer. This work proposes an innovative solution using anArtificial Neural Network (ANN) controller to dynamically adjust the current in response to the fluctuating coupling factor. By leveraging machine learning, the ANN can predict and compensate for misalignments, ensuring stable and efficient power delivery in a multi-pad charging station. Simulations conducted using the Simscape Power Systems module in Simulink demonstrate the effectiveness of the proposed ANN-based tuning mechanism in maintaining optimal charging performance. This approach offers a promising path toward enabling large-scale, autonomous drone recharging stations, paving the way for continuous and efficient drone operations across various industries.