Based On The Video Inverted Swing-up Control System

As the development of computer science and image processing, Computer vision information can be used in feedback of a system. One of control methods based on visual servoing was be given. That means according to the principle of computer vision, the system process the image which was given by video camera quickly, and output the message of the object as fast as well. So the closed loop position system can be realized. Detection of moving-targets in complex background is a crucial aspect of Image Processing, and it is also an important application of computer vision in practice. As a piece of image include a great plenty of complex data, it will spend a lot of time for us to detect the moving targets in the real-time image. We must improve our processing method to reduce the processing time and error.The control of an inverted pendulum is a known and well-understood task. It's structure is simple and can be easy to understand. The parameter and the structure of the system can be modify easily.The inverted pendulum is high-degree unstable multi-vector and nonlinear system. Reach on inverted pendulum have a important practice background.In this paper we present a new approach of controlling an inverted pendulum with a vision video camera. As we known, the sampling rates of CCD camera is 20-100 frames per second. This sampling rate results in a very sluggish control of a pendulum even in the ideal case when encoders are used. How to design a robust intelligence method which can control the system better in the condition of low sampling rate and bad quality of signal is the key. In this paper we analysis the characters of image and the features of disturb firstly. We give a method that can reduce the contribution of disturb. And then the split of threshold quantity and change detection mask is applied in how to detect the position of moving object separately. We design a method based on human-imitating intelligent control which can swing up the inverted pendulum with a video camera.