The pure pursuit model is widely used in the path tracking control system of agricultural machinery. However, system oscillations occur easily at high speed. Aiming to improve the stability of the trajectory tracking system of automatic rice transplanter, a path tracking method based on fuzzy control was proposed to adjust the lookahead distance of pure pursuit model. Firstly, the kinematic model of the transplanter was established by considering the automatic steering system as firstorder model. Secondly, by linearizing the kinematic model, the stability condition of pure pursuit model was analyzed while tracking a straight line. Based on the stability condition, fuzzy control model was established to adjust the lookahead distance of pure pursuit model, with taking the speed and lateral deviation as inputs and the lookahead distance as output. To construct the experimental platform, automatic transformation was carried on with Yanmar VP6E rice transplanter. Finally, to verify the availability of proposed method at different speeds, the experiments were carried out at high speed of 1.0m/s and low speed of 0.3m/s. Compared with the traditional pure pursuit model with fixed lookahead distance, the method proposed can avoid oscillation at high speed and track the target path quickly at low speed. It was proved that this method can effectively improve the stability of the pure pursuit path tracking system.