This paper proposes an adaptive feedforward control combined with a PI controller to be applied to a Permanent Magnetic Synchronous Machine (PMSM). The adaptive technique is realized by using an Extended Kalman Filter which provides for the estimate of paramteres and states necessary for the adaption. The estimated inductivities and the estimated speed are used in a feedforward control to improve the reference variable attitude. The impact of the feedforward control on the control will be examined and the PI controller will be set up.