Abstract
<div class="section abstract"><div class="htmlview paragraph">This work proposes a unique control method consisting of ameliorated with reinforcement learning renewal module. The combined fuzzy logic and reinforcement learning regime is utilized to promote robust energy management control in complex working conditions. The coupled optimization proposition tackles unforeseen disturbance by two-pronged approach, with fuzzy logic analyzing backbone power contribution schemes while reinforcement learning takes responsibility for improving a higher efficiency strategy. The vehicle dynamic parameters and energy map are co-modeled through learning extrapolation function. Fuzzy rule undergoes efficient feedback revival via modulating factors driven from multi-objective RL reward computation. Meanwhile, reinforcement learning system leverages adaptive fuzzy representation that generalizes coordination potential vectors, effectively extends exploration quality compared to vanilla learning strategy. To this end, this work effectively considers traction condition, separate wheel control, collaborative effort as well as battery aging into holistic management deduction through renewal regime. To resolve transient contradictions stemming from fuzzy structure regulation, a duet pattern regenerating method is developed, which transfers mechanical and electrical considerations into a unified competence factor. The performance of this system is delineated through numerical simulation and experimental results through Simulink. The results provide multi-faceted insights for the field of adaptive control with application to nonlinear dynamics in addition to emergent influence occurring in more advanced models.</div></div>