Study In Design And Experiment Of Inverted Pendulum Servo System

The problem of balancing an inverted pendulum by moving the pivot horizontally on a cart is commonly used in control education to provide a dramatic experiment and to illustrate the difficulties in controlling a plant having unstable open-loop poles. Inverted pendulum servo system has several attractive features not only from a pedagogical standpoint but also from a research standpoint. As a complex, time-varying, nonlinear, strong coupling, natural unstable and higher order system, its stability control is a typical problem of the theory and application of control.The contents of this paper consists of the design and realization of motion control model for the inverted pendulum servo system, modeling and emulating of the inverted pendulum. Furthermore the experiment of balancing a single inverted pendulum was accomplished, the pendulum was maintained at upright position.The motion control model is based on digital signal processor, which develop the specialties of the system such as speediness and real time, and improve the speed and precision of data processing and calculation.On the assumption of having no friction, single, double and triple inverted pendulum is modeling and emulating with LQR-based linear control strategy partly. Furthermore, the friction influence to the linear model of inverted pendulum was analyzed, the common style of inverted pendulum model with friction influence is given, the principle of modifying the strategy with friction influence is discussed.The control program is developed with C language, and it can accomplish the functions of digital transfer, strategy disposal, control output and communication interface. The software of PC is developed with VC, and it supply a user interface with the functions of inspect, control, and debug online.Debugging and experiment of single inverted pendulum is accomplished, function of following and positioning was realized. Results of the experiment shows that the servo system having stated characteristics of transient, steady state and robustness. The work presented here, not only can be used as a pedagogical experiment for modern control theory, but also can benefit practical problems with similar model such as the study of bipedal robots.