For validating the handling characteristics of a vehicle, data on front and rear axle forces are very insightful. The traditional approach to vehicle testing involves physical sensors, which can be expensive, time-consuming to install, and prone to measurement errors. Model-based virtual sensing offers an alternative approach, using a combination of easier-to-measure sensor data and an accurate numerical model, in order to estimate physical quantities of interest. This work presents a framework for developing virtual sensors based on a 15 Degrees of Freedom vehicle model. The effectiveness of the proposed technique is evaluated through simulations and the results demonstrate that the virtual sensor can estimate the vehicle’s axle loads with accuracy and robustness.