An intelligent vehicle control system based on visual path recognition was designed to recognize various complex paths quickly and stably.MK60DN512ZVQ10 controller was used as the core of the intelligent car control system,and OV7725 vision sensor was used to obtain the binarization image of the track.The edge and center line of the track were extracted by image processing,and a recognition algorithm for complex path such as roundabout was proposed.The incremental encoder was used to measure the realtime speed,and the PID control algorithm was used to control the shifting of the steering gear and the speed of the drive motor.By adjusting the height of the vision sensor and reasonable layout of the car body structure,the path recognition capability and the stability of the intelligent vehicle were improved.The experimental test results show that by using the control system,the intelligent vehicle can run quickly and stably on the track and correctly recognize various paths when the speed is about 2.4 m/s.