The move toward Electric Vehicle (EV) usage is rapidly gaining traction as a response to improving environmental conditions. As the adoption of EVs accelerates, several existing concerns and limitations need to be addressed. One of the main challenges affecting the widespread shift to EVs is the uncertainty surrounding the remaining range that a vehicle can travel. Accurately predicting this remaining range depends on various factors that must be considered. This can be achieved by integrating essential vehicle and trip data to train the predictive model effectively. While many existing models often lack the necessary granularity for reliable accuracy under different driving conditions, this approach seeks to implement federated learning. This method enables EVs to collaboratively improve predictive models, ensuring accurate forecasts while protecting sensitive information. Additionally, the integration of blockchain technology with federated learning guarantees that all updates are recorded transparently and maintain data integrity. Effectively predicting the residual range the EV can travel can significantly alleviate drivers’ range anxiety. This proposal aims to provide accurate and secure predictions of the remaining range an EV can travel.