Published Paper Details:

T Lithium-ion batteries remain foundational to electric vehicle propulsion and stationary energy storage, yet the persistent tension between fast charging throughput and long-term electrochemical health continues to constrain their practical deployment. Conventional constant-current constant-voltage protocols, while computationally inexpensive, impose rigid current and voltage boundaries that fail to adapt to progressive cell degradation, accelerating solid-electrolyte interphase growth and lithium plating over repeated charge cycles. Advanced control paradigms - spanning model predictive control, side-reaction overpotential regulation, and increasingly, deep reinforcement learning - have emerged as promising alternatives capable of balancing charging speed against battery longevity. Among these, lifelong or continual reinforcement learning frameworks represent the most ambitious frontier, enabling charging agents that continuously adapt their policies as cells age without discarding prior learning.. A structured literature synthesis table is presented to facilitate direct cross-study comparison across topology, method, key contribution, and identified limitation.