Reinforcement learning in electric vehicle energy management: a comprehensive open-access review of methods, challenges, and future innovations

Electrification of transport is accelerating worldwide, raising new challenges for energy efficiency and control in electric vehicles. Reinforcement learning has emerged as a promising data-driven approach to address the complexity of real-time energy management. This review presents a structured synthesis of open-access research published between 2016 and 2024 on the application of reinforcement learning methods to electric vehicle energy optimization. The study formulates four guiding research questions to analyze types of learning algorithms, evaluation criteria, system-level constraints, and practical implementation aspects. Key contributions include a comparative mapping of reinforcement learning techniques—such as Q-learning, deep deterministic policy gradient, twin delayed deep deterministic policy gradient and soft actor-critic—their applicability to electric vehicle control scenarios, and the identification of current research gaps and deployment challenges. The findings aim to support researchers and engineers in selecting suitable reinforcement learning strategies for efficient and scalable electric vehicle energy management.