Robust-enhanced tube-based robust model predictive control for lateral stability of steer-by-wire vehicles

The steer-by-wire system eliminates mechanical linkages, enabling flexible front-wheel steering and improved vehicle handling stability. To address system uncertainties and external disturbances, this paper proposes a tube-based robust model predictive control framework. The main contributions are as follows: (1) the design of a closed-loop feedback gain K that ensures asymptotic stability of the error system and minimizes the deviation between nominal and actual vehicle dynamics, obtained offline using Linear Matrix Inequalities and the Schur complement to accommodate high-dimensional systems; (2) a systematic nominal–actual system analysis and an error-reduction strategy to improve overall control accuracy; and (3) integration of these strategies into the tube-based robust model predictive control framework to guarantee robust lateral and roll stability under bounded disturbances. Simulation results demonstrate that, compared with conventional variable-weight model predictive control, the proposed method significantly reduces sideslip and roll angles while maintaining control accuracy and robustness, highlighting its effectiveness for steer-by-wire vehicle control.