Abstract

Accurate axle mass estimation is critical for vehicle control and maintenance, but conventional GPS- or brake-sensor- based methods are impractical for rear-wheel-drive electric vehicles (EVs). This study presents a CAN-based method that requires no additional hardware. Slip ratio is estimated based on the difference between wheel speed sensors on the driven axle (with slip) and the non-driven axle (without slip). The calculated slip ratio is then applied to a linear tire model to estimate the rear axle load, considering motor torque and tire dynamics. Load transfer during acceleration is then compensated to obtain static axle mass. Experiments on a Hyundai Ioniq 5 under rainy conditions produced 18 datasets across two speed conditions and three loading scenarios. Results show estimation errors within ±10% under all conditions, with improved accuracy through load transfer compensation. The method demonstrates strong potential for integration into autonomous driving, tire wear prediction, and prognostics and health management (PHM).