The rapid growth of semiconductor and display manufacturing highlights the demand for fast, precise motion stages. Advanced systems such as lithography and bio-stages require accuracy at the μm and nm levels, but linear motor stages face challenges from disturbances, model uncertainties, and measurement noise. Disturbances and uncertainties cause deviations from models, while noise limits control gains and performance. Disturbance Observers (DOBs) enhance performance by compensating for these effects using input–output data and a nominal inverse model. However, widening the disturbance estimation bandwidth increases noise sensitivity. Conversely, the Kalman Filter (KF) estimates system states from noisy measurements, reducing noise in position feedback, but it does not treat disturbances as states, limiting compensation. To address this, we propose an Augmented Kalman Filter (AKF)–based position control for linear motor stages. The system was modeled and identified through frequency response analysis, and DOB and AKF were implemented with a PIV servo filter. Experimental validation showed reduced following error, jitter, and control effort, demonstrating the improved control performance of the AKF approach over conventional methods.