The performance of battery management systems relies on the core temperature estimation, which is one of the major technical bottlenecks for electric vehicles. Aiming to tackle this problem, a lumped-parameter two-state thermal model for lithium-ion batteries is established in this paper. Then, this thermal model is coupled with a temperature-dependent second-order RC equivalent circuit model to form an electro-thermal model for lithium-ion batteries. Using the proposed electro-thermal model, an adaptive estimation algorithm based on joint Kalman filtering is proposed for battery core temperature estimation, considering the heat transfer condition variations between the battery surface and ambient media. The verification results show that the proposed algorithm enhances the temperature estimation accuracy, compared to the results obtained directly from the electro-thermal model. Besides, the verification results demonstrate the high adaptability of the proposed algorithm, that is, it is robust to variations in ambient temperature, as well as variations in thermal resistance between the battery surface and ambient media. Considering the influences of temperature and SOC on the thermal generation rate of the battery, an electro-thermal model for lithium-ion batteries is established. This model is composed of a lumped-parameter two-state thermal model and a temperature-dependent second-order RC equivalent circuit model. An adaptive core temperature estimation algorithm based on joint Kalman filtering is proposed, considering the heat transfer condition variations between the battery surface and the ambient media.