• Proposes a robust Tube-MPC trajectory tracking control method for 4WIS vehicles on intermittent icy and snowy roads, addressing time-varying adhesion coefficients and cornering stiffness uncertainties.
• Integrates a Bi-directional LSTM neural network for online estimation of tire cornering stiffness, enhancing the accuracy of the nominal MPC model under varying road conditions.
• Combines Tube-MPC with robust Sliding Mode Control to handle road irregularities, improving trajectory tracking accuracy and robustness compared to standard Tube-MPC.
• Experimental results demonstrate superior trajectory tracking performance under challenging road conditions, providing a theoretical foundation for future vehicle stability and control studies.