Electric vehicles (EVs) operate under diverse environmental conditions, which significantly influence the performance and lifespan of their batteries. This study investigates the combined impact of charging rates and ambient temperatures on the degradation of lithium iron phosphate (LFP) batteries commonly used in EVs. A novel hybrid machine learning (ML) approach is employed using an extreme gradient boosting random forest (XG-RF) model to estimate the state of health (SOH) of battery, factoring in key parameters such as charge cycles, voltage, and surface temperature. The analysis evaluates battery cycle life across varying charge rates (C/20, 1C, 2C, and 3C) and three different temperatures (5 °C, 25 °C, and 35 °C). The results highlight the superior cycle life of LFP batteries, especially at moderate temperatures, achieving up to 5293 cycles at a 25 °C temperature and a C/20 charge rate. These insights provide critical guidance for enhancing battery performance in EVs and contribute to advancements in battery health monitoring and optimization strategies.