The article presents a method to accurately estimate the state-of-charge (SoC) of a multi-cell battery pack used in electric vehicles. The researchers developed a filtering process to select a representative cell, improved the battery model to account for uncertainties, and created an adaptive SoC estimator. This approach allows for precise prediction of SoC for each cell in the battery pack, enhancing safety during operation. The results show that the estimation errors for voltage and SoC of all cells are less than 3%, even with initial errors in the estimator.