The present paper proposes a novel fast and precise positioning strategy using an iterative learning control. The authors have previously proposed an effective positioning algorithm by applying the command shaping technique using an FFT-based inverse model of a closed loop system. On the other hand, the modeling error and/or variation of the plant system causes deterioration of the positioning performance due to the lack of robustness of the controller with respect to these errors. In the present research, in order to ensure the above-mentioned robustness, an iterative learning control technique in the frequency domain is introduced, in which the position command is successively shaped to satisfy the positioning specifications. The effectiveness of the proposed control strategy has been verified experimentally using prototype machine tools.