Electric vehicles (EVs) have been heavily used to minimize the worldwide pollution. Battery storage system is the most important and expensive system in these vehicles. An accurate battery management system (BMS) must be applied to monitor and control the battery states. From these measurements, the residual useful life (RUL) is estimated to avoid any further safety issue which can destroy the battery system or vehicle, or harm the passengers. Many features are required to be measured to estimate the battery states and then estimate the RUL such as battery voltage, current, and temperature according to the time. These measurements are taken using embedded system, and the measured data are stored in data files. Artificial neural networks (ANNs), long short-term memory (LSTM), support vector regressors (SVRs), random forest (RF), and boosting methods have been implemented to estimate the battery RUL. Moreover, many optimization algorithms such as particle swarm optimizer (PSO) and whale optimization algorithm (WOA) are integrated with the extreme learning machine (ELM) to estimate the battery RUL. The root mean square error (RMSE) is considered the main factor that has been used to judge which algorithm is more accurate than the other. It is found that both PSO-ELM and WOA-ELM supersede the machine learning tools, where the RMSE values are 1.46% and 1.51%, respectively. These values are 2.24%, 2.25%, 2.74%, 2.84%, and 3.56% LightGBM, random forest, AdaBoost, XGBoost, and CatBoost algorithms, respectively.