Harnessing Symbolic Regression: Optimizing Distance Estimation for Wireless Power Transfer in Underwater Vehicles

Symbolic Regression (SR) has emerged as a powerful tool for optimizing wireless power transfer (WPT) systems, particularly for autonomous underwater vehicles (AUVs). This study explores the application of SR in estimating the distance between coils in WPT systems, leveraging both simulated and experimental data to achieve low prediction errors. By employing a genetic algorithm, SR identifies analytical formulas that represent the input-output relationship, combining machine learning and analytical modeling. The results demonstrate that SR not only enhances predictive accuracy but also provides insights into the physical principles governing WPT systems. These advantages are critical for addressing the unique challenges of underwater environments, such as seawater conductivity and alignment precision. By enabling more efficient and reliable power transfer, SR contributes to advancing underwater WPT technologies, ensuring continuous operation of AUVs and reducing reliance on surface support vessels. This positions SR as a transformative method in the development of sustainable underwater exploration and maintenance systems.