Minimizing Impacts of Road Adhesion Perturbations: Uncertainty Allocation for Control Synthesis of X-by-Wire Chassis

Model-based optimal control has been widely adopted for vehicular stability enhancement. However, existing schemes still suffer from unmodelled external disturbances such as road adhesion variations, which leads to significant performance degradation. The recent progress in X-by-wire chassis brings promising solution to address this challenge. Utilizing independent steering along with distributed driving systems, this paper proposes a vehicular control programs that actively distribute tire forces in both longitudinal and lateral directions to minimize the impact of external parameter perturbations. Firstly, an adhesion disturbance modeling approach integrating composite slip and single-wheel reachability analysis is developed to accurately characterize the feasible region of tire forces under adhesion disturbances. Secondly, based on the disturbance propagation mechanism, the sensitivity differences of various tire force distribution strategies to key vehicle states are systematically analyzed, leading to the proposal of an active distribution criterion with minimal disturbance response sensitivity. Furthermore, a stochastic model predictive control (SMPC) framework integrating state covariance constraints and probabilistic constraints is designed, which improves control accuracy while significantly suppressing the propagation of state uncertainty. Simulation results demonstrate that the proposed method effectively reduces state fluctuation ranges under typical longitudinal and lateral conditions, thereby enhancing vehicle handling stability and robustness under adhesion uncertainty.