In this experiment, a smart Battery Management System (BMS) 4S is designed and developed, and the system is validated in a laboratory environment. The proposed system integrates advanced technologies to improve the monitoring, control, and overall performance of batteries in various electric vehicle applications. The real-time monitor of the smart BMS should take care of the key battery parameters such as voltage, current and temperature and also offer the ability to monitor and adjust the operation and safety of the BMS better. Furthermore, the data generated from the BMS sensor is utilised to estimate the State of Charge (SoC), a critical aspect of battery management, which is achieved through a sophisticated algorithm. These algorithms were evaluated in terms of their performance as compared to traditional approaches, the Coulomb Counting (CC) method. We find that regular techniques are invariably outperformed by machine learning methods, with an average estimation error of less than 1 \(\%\) . Finally, we also take a look at the advantages and drawbacks, and to what extent machine learning is suitable for SoC estimation, and indicate where future research should focus. Importantly, these results have implications for the development of more effective battery management systems, which can be used to ease the entry into battery management systems, increase battery performance, safety and life.