In this article, a disturbance estimation–based robust generalized predictive control approach is proposed for angle demand control of a steer-by-wire system. It consists of a generalized predictive controller and a nonlinear disturbance observer. First, the nonlinear disturbance observer is used to estimate the effects of the lumped disturbances, including model uncertainties, unmodeled dynamics, and external disturbances. Second, the estimate of the nonlinear disturbance observer is integrated into the generalized predictive controller design for front-wheel angle output prediction. Then, an explicit analytical form of the disturbance estimation–based robust generalized predictive controller is derived. In addition, rigorous stability analysis proves that the closed-loop system under the proposed control is stable. Finally, the developed control approach has been downloaded into a steering control unit and tested in real-world conditions using a vehicle test bench to fully realize steering by wire in engineering practice. Extensive simulations and experiments have been performed, and results show that the proposed control approach outperforms the comparative controllers regarding disturbance rejection, robust tracking performance under model uncertainties, and disturbances.