In order to improve the path following accuracy and driving stability of intelligent vehicles, a path following method based on improved pure tracking is proposed. Firstly, the path following model based on pure tracking is established, and the control effect of the pure tracking model under the condition of initial error is analyzed. Secondly, the heading error is used as the feedback variable to compensate the pure tracking method, and the quasi fuzzy method is used to realize the dynamic adjustment of the compensation weight with the lateral error and the heading error as the input and the compensation weight as the output. Finally, the control parameters are selected through the real vehicle test, and the control performance of the proposed method is verified. The results show that the method can eliminate the errors quickly under different initial errors and different target driving speeds and ensure the driving stability, and the method has better path following effect and spends less compute time.