For the existing problems that system lag, friction, and uncertainty factors in linear permanent magnet synchronous motor drive XY table, a combinative control method of generalized predictive controller (GPC) and disturbance observer (DOB) is put forward. Firstly, through GPC the parameters of system are identified, controlled autoregressive integral moving average model (CARIMA) is established, and then rolling optimization, local optimum of system is sought. Finally feedback is corrected, and system can be calibrated continued. The control lag problem of XY table system is solved. The tracking ability of system is improved. System parameter uncertainty and various nonlinear disturbances are eliminated or weaken by DOB. Robustness of the system is improved. The simulation and experimental results show that the proposed scheme is effective. The tracking performance and robustness of system are improved obviously. So the machining precision of the system is improved.