This paper proposes and describes a control method that compensates time delay in a system by predicting the future states of a plant. The proposed method contains a state predictor of future values of the system. Furthermore, the method considers uncertainty in the model in terms of modeling errors, which might cause the predicted value to degrade the performance of the system or even destabilize it. The disturbance in a remote plant is explicitly considered and is compensated for by placing a feedback controller on the remote side. This paper describes the proposed control method based on model-based predictive control that considers modeling error. The stability of the proposed system is analyzed, and the performance of the method is evaluated through a comparison with conventional methods using numerical simulation and experimental results.