To tackle the problem of high performance robust control of induction motor under parameter disturbances, an unbiased speed control method for induction motors based on model predictive control technique and disturbance observer was proposed. Based on the cascade control structure, the inner loop model predictive torque controller and the outer loop robust speed controller were then given. An improved continuous control set model predictive controller (CCS-MPC) was proposed as the inner loop torque controller on the premise of considering the minimum loss under motor current and voltage constraints and the variation of parameters with speed; the stabilized MPC method was combined with the discrete time domain disturbance observer (DOB), and the outer loop robust speed controller was designed under the premise of considering the influence of parameter uncertainty and the load torque and torque limit of the motor, which provided reference torque for the inner loop torque controller; the robustness of the proposed speed controller was illustrated by the corresponding theorem and lemma; physical experiments were designed to verify the proposed controlling method. The results show that under the condition of considering the torque limit of the motor, the proposed method has good transient state and steady state response, and it has good robustness under load torque disturbance.