<p>The pressing challenges posed by fossil fuel dependency and its detrimental effects on the environment have become a catalyst for the global transition to sustainable energy solutions. Among these, electric vehicles (EVs), driven by advancements in battery technologies, stand out as a pivotal solution to simultaneously tackle the energy crisis and mitigate environmental degradation. Accurate battery modeling and state estimation are critical for optimizing energy usage, extending battery life, lowering operational costs, and ensuring the reliable and safe functioning of electric vehicles (EVs). However, reliable state-of-charge (SOC) estimation remains challenging due to the dynamic and nonlinear characteristics of batteries. Traditional methods have frequently struggled to deliver accurate and reliable SOC predictions, particularly when faced with noisy input data and the complex, fluctuating patterns inherent in SOC models. This study introduces and evaluates a hybrid deep learning model that integrates recurrent radial basis function (RRBF) networks with fuzzy c-means clustering, designed to improve the accuracy of state-of-charge (SOC) estimation in batteries. The findings reveal a significant improvement, with a 0.43% reduction in prediction error on noisy datasets, showcasing the model’s superior performance compared to traditional methods. This contribution not only enhances battery management systems but also supports the advancement of more efficient and sustainable electric vehicle technologies.</p>