State-of-X estimation for lithium-ion batteries in electric vehicles: A comprehensive review of methods, challenges, and emerging trends

An essential component in improving the efficiency, safety, and performance of electric vehicles (EVs) is the battery management system (BMS). The BMS plays a critical role in controlling and monitoring the battery to ensure optimal operation under various driving and environmental conditions. Central to this function is the accurate estimation of key battery states, commonly referred to as the State of X (SoX), which includes the state of health (SoH), state of charge (SoC), state of power (SoP), state of energy (SoE), and remaining useful life (RUL). Hybrid approaches combining data-driven and model-based methods offer robust SoX estimation, yet their practical use for multiple parameters in EV BMS remains limited. This review aims to address this gap by providing a comprehensive and systematic analysis of popular machine learning algorithms used in battery SoX. The review contrasts the advantages and limitations of different ML-based methods and explores their application to SoX estimation. Feature extraction techniques employed in recent studies are also discussed. Furthermore, this review offers a thorough examination of hybrid techniques, focusing on cutting-edge methods, executions, precision, benefits, limitations, and contributions. To aid future research and practical deployment, a multi-criteria decision matrix is introduced to map estimation requirements to suitable method categories for each battery state. Finally, the review outlines future research directions and emerging opportunities aimed at advancing hybrid and co-estimation frameworks. These insights are intended to support EV developers and automotive industry in enhancing BMS capabilities—contributing to reduced carbon emissions and the achievement of global sustainability goals.