With the development of science and technology, there are more and more types of sensory sensors of the robot, among which the visual sensor becomes an important part of automatic walking and driving. A typical area of view is an autonomous intelligent navigation system. Image processing technology has been quite developed for different visual technologies, while image understanding technology is still very backward. Machine vision requires multiple operations and can only recognize a few simple goals in a structured environment. The basic element of the vision sensor is the camera tube or CCD, which can focus automatically. However, the price, usage and mode of use of the CCD sensor are not dominant, so it is a practical and effective method of taking into account the proximity sensor in the system, which does not require a clear picture, but only unpleasant feeling. To perform the function of automatic targeting and avoidance of obstacles, the robot must perceive the wires and obstacles that are equivalent to the visual function for the robot a.The obstacle avoidance control system is based on the avg-auto-vehicle guide system. Based on the avg-auto-vehicle guide system, the smart vehicle can automatically identify the route, evaluate and automatically avoid obstacles, and choose the right route. Use sensors to sense routes and obstacles, make judgments and perform actions appropriately