The Fuzzy Control Of Inverted Pendulum System

As the research hotspot of control field, the inverted pendulum device is thought as typical test-equipment in the auto control theory, and it is also typical physical model in the teaching and scientific research of controlled theory. Moreover its controlled way and idea have abroad use to deal with common industry course. To its research may not only solve problem of controlled theory, but also will carry through synthesize use to three mostly basic subject :mechanics,mathematics and electricity of controlled theory involved. Furthermore, the inverted pendulum may provide a bridge from controlled theory to practice. Thus, the inverted pendulum system is a test-equipment of rather unearthly in the controlled theory research. The text discusses exploitation course that the self-adapt fuzzy control of inverted pendulum system which is assisted with MATLAB language. To fuzzy theory and self-adapt neural network's amalgamation, it is availability attempt and innovation to the controlled aspect of inverted pendulum system.Above all, this task set down feasible system design direction from design demand of inverted pendulum system. Form system exploitation flat roof, under Windows circumstance, MATLAB language assistant design have several characters: program efficiency high,sentence simple,using effect and convenient and so on. Next, after system model of dolly inverted pendulum problem carry through dynamics analysis and predigest, using fuzzy illation system of basing on self-adapt network ANFIS research the inverted pendulum's controlled problem. In the fuzzy control choice Takagi-Sugeno model to the controller's analysis and design is cooperate with linearity system theory. Again, according to practice control demand, the range of ensure state variable and controlled variable, at a direction of state space equality choose some data point. When obtaining plenty data, self-adapt neural network fuzzy control system ANFIS come into being subjection degree function and fuzzy rule, namely come into being fuzzy controller.In the end, Simulink that is provided imitate circumstance by using MATLAB system carry through imitation for system. Ratiocinate system that is designed directly lead to imitate circumstance, and ratiocinate system that is designed will be proved and tested by imitation. So you say fuzzy rules that is system stabilization and dynamic track ability preferably, so control result comparatively ideal.