AI-driven battery intelligence for energy management in electric vehicles: Strategies to enhance safety, performance, and lifetime — A review

Advanced battery intelligence is becoming increasingly important for the safe, effective, and sustainable operation of electric vehicles (EVs). This paper examines the integration of artificial intelligence (AI) methods into battery management systems (BMSs) and their cooperation with energy management systems (EMSs), including machine learning (ML), deep learning (DL), and reinforcement learning (RL). We focus on how AI-driven state estimates, such as the state of charge (SoC), state of health (SoH), and remaining useful life (RUL), are essential inputs for EMS decision-making in various applications, including thermal management, power deployment, and predictive charging. To demonstrate their functions in fleet-level optimization, we discuss predictive maintenance, and real-time monitoring, recent advancements in cloud-based BMS architectures, digital twins, and data-driven control algorithms. Along with outlining important future possibilities, including lightweight AI models, cross-fleet generalization, and explainability, this article addresses the present challenges with interoperability, standards, and computing limits. The transformational potential of AI-enhanced EMS–BMS coordination in influencing next-generation EVs performance, battery longevity, and sustainable transportation is highlighted.