Neural Network Fuzzy Control In The Application Of Inverted Pendulum System Research

Inverted pendulum is an absolutely unstable, high order time, multivariable andstrong coupling nonlinear system, it unifies the robot technology intelligent control andcomputer control organically. As a kind of typical experimental device, Inverted pendulumsystem can verify control theory effectively. It has a wide range of practical guidingsignificance to complex controlled object control in various industries. This paper treatshigh solid double inverted pendulum as control object, adopts the Newton’s mechanicsmethod and Lagrange equation method, and conduct mathematical modeling aboutprimary and secondary inverted pendulum system. And analyzes its controllability andstability. It based on the mathematical model of the double inverted pendulum system,adopted LQR theory controller, fuzzy controller to control the double inverted pendulumsystem. Then it based on fuzzy control theory and neural network control theory, thispaper proposes a fuzzy control method based on neural network. This kind of controlmethod through the optimal control method to reduce the number of input variables of theinverted; Using hybrid algorithm combined with the least squares algorithm and thegradient descent method. It uses obtained input and obtained output data sets, constantlyrevised the shape of membership function of fuzzy controller and the control rules, getdesirable output data. Simulation experiments show that the method can control the doubleinverted pendulum system effectively, and robustness in the system is good. And itscontrol effect is better than that of general LQR controller. At last, Using the method oflinear quadratic optimal control, perform an real-time control experiment and ananti-jamming control experiment about double inverted pendulum.