The presented research paper explores the expanding field of electric vehicles and hybrid automobiles, focusing on the critical issue of accurate state of charge (SoC) prediction, one of the key parameters in the analysis of vehicle range. SoC prediction not only depends on the internal structures and functions of the vehicle and Li-Ion battery parameters but also external field contributors such as temperature, friction, etc., making it a complex parameter to understand and predict. This study conducts an in-depth comparative analysis of three distinct machine learning algorithms, namely Adams Optimizer, RMSPRop, and Adagrad. These machine learning models were trained under identical conditions post extensive feature engineering on the dataset. The model comparison was meticulously conducted via different evaluation metrics, and the results and further approaches are presented in this paper.