In order to improve the accuracy of the prediction of the track, and strengthen the impact of the attack on the target, an improved grey residual model is proposed to predict the target track. The method of grey GM(1,1) model with support vector regression combination, obtained by predictive value and residual sequence of the original track data to establish the grey model, the improved residual model is used to correct the predictive value by using the nonlinear fitting ability of support vector regression, to a certain extent overcome the GM(1,1) model the model defects, the prediction accuracy can be improved. Using the grey GM(1,1) model, the grey residual GM(1,1) model and the improved grey residual GM(1,1) model, 3 methods are used to predict the same track and to compare the predicted results. The results show that the improved model can predict the track accurately with the large variation of track data, which has a high theoretical and practical reference value.