The distributed driving system increases the vehicle dynamic control possibility. The driving torque of wheels can be adjusted to maintain vehicle stability based on the yaw rate and vehicle sideslip angle generated from the reference model. To improve vehicle stability in different road conditions, an adaptive model predictive control (AMPC) method is provided based on the vehicle sideslip angle phase plane and different force areas. The weight factor of the ideal state parameter is adjusted based on the stability indicator while the stability control method is determined based on the tire force state. The provided method is validated by a double-lane changing (DLC) maneuver on a high and low cohesion road simulated in the hardware in the loop (HIL) platform. According to the simulation, the controller can improve the vehicle stability in the typical working conditions, thus, improving the vehicle handling performance and stability.