In this paper, a new hybrid method for controlling a wheeled robot is introduced. Model predictive control (MPC) is the main controller and a fuzzy controller is used as a compensator. The wheeled robot is a nonlinear, multi-input–multi-output system that requires new and combined methods for precise control. In order to stabilize the system the appropriate control input is set, and at the same time, attention is paid to the reference signal tracking. In the simulation section, several different scenarios are applied and parameter uncertainties and their effects on the controller’s performance are evaluated. The simulation results show the success and efficiency of the proposed method.