This research introduces a parallel control architecture aimed at enhancing driving stability for wheel-legged vehicles by addressing instability in the yaw and roll planes during high-speed sharp turns. The aim is to fully utilize the vehicles’ high mobility and passability. Initially, a 15-degrees-of-freedom model of the entire vehicle is developed. The proposed approach involves implementing a parallel controller to decouple yaw and roll motions. The drive system utilizes a model predictive control (MPC) algorithm to regulate vehicle roll motion, while the leg system employs a virtual model control (VMC) algorithm to govern active roll control. Ultimately, simulation analysis is conducted. The findings demonstrate that the control approach rapidly and efficiently tracks the desired value. Additionally, the study shows the stabilizing impact of active roll on the yaw and roll planes of wheel-legged vehicles.