Intelligent agricultural vehicle can significantly improve the production quality and efficiency, reduce environmental pollution, and take advantage of the agricultural resources efficiently. So an automatic guidance control system in agricultural vehicle was implemented by integrating the hardware and software.The hardware included sensors, actuators and CAN-Bus.The software was developed with multithread programming technology in Visual Studio 2005 platform based on Windows operating system. Openloop transfer function model of the heading control system and the lateral control system were established via the model of the vehicle motion and the steering system. Model of the vehicle motion was established according to the kinematic characteristics, and model of the steering system was identified according to the closedloop response curve. Then the heading controller and the lateral controller were designed via PID control theory. Tests results showed that the hardware and software system were running stability. With the initial heading deviations of -86° and 84°, the adjusting time were both in 2s, the turning precisions were both in 1°. With the initial lateral deviations of 0.7m and 1.2m, the maximum values of the lateral deviation were 10.4cm and 9.2cm respectively, the mean values of the lateral deviation were 6.4cm and 3.5cm respectively, and the standard deviations of the lateral deviation were 2.6cm and 1.7cm, respectively. So the proposed automatic guidance control system can make the vehicle track the desired path smoothly and steadily with a centimetrelevel tracking precision.