This paper presents an advanced methodology for SOC estimation by integrating Recursive Least Squares (RLS) techniques with Adaptive Extended Kalman Filter (AEKF). The proposed methodology aims to mitigate the challenges associated with fluctuating battery parameters and varying noise characteristics over time, which can significantly impact the accuracy of SOC estimation. By dynamically adjusting to evolving system dynamics and noise statistics, the proposed approach exhibits enhanced robustness and accuracy compared to traditional techniques. The proposed methodology assumes that battery parameters, including internal resistance, capacitance, and noise information, undergo variations over time. To address this assumption, two distinct online identification algorithms for parameters and noise information are introduced. Specifically, the RLS algorithm is utilized to ascertain resistance and capacitance values. Process and measurement noise covariance is also estimated based on an iterative noise information identification algorithm. Subsequently, all updated values are incorporated into the EKF. The results demonstrate that the RLS-AEKF approach achieves higher accuracy than the EKF. The results based on Fast Urban Driving Schedule (FUDS) and Urban Dynamometer Driving Schedule (UDDS) working current profiles validate the effectiveness of the proposed approach in enhancing SOC estimation accuracy under realistic operating conditions.