Inverted Pendulum To The Study Of Intelligent Control Systems

Inverted pendulum which is a typical fast, multivariable, non-linearity, strong-coupling and naturally unstable system is a classical control problem of balancing stick vertically on a linear track cart. During its control process, it can reflect many crucial questions in the control theory, such as calm question, non-linear problem, robust question as well as tracking question and so on. Then it is the favorite problem in the control theory. The research on inverted pendulum system has the profound significance on theory and project applications. The correlative scientific research achievement has already been applied to astronautics science technology and subjects of robot and so many domains.This paper analytically investigates the structure of inverted pendulum and the latest relative research productions at home and abroad. On this basis, the non-linear mathematical model of inverted pendulum is formulated. Then, in the vicinity of the equilibrium point of it, the model is linearized and the state-space equation is obtained. This paper studies theory foundation of fuzzy control and neural network control. And it analyses shortcomings of both each and the inevitability of their combination. On this foundation, a T-S fuzzy neural network controller based on T-S fuzzy model is designed.Towards to single inverted pendulum, using the learning capability of nerve network to train membership function of the fuzzy controller, establishes a T-S fuzzy neural network controller to control the inverted pendulum through the Adaptive Neuro-Fuzzy Inference System (ANFIS). Exceptionally this paper designs a fuzzy control system. The results of simulation experiment of the two control systems prove that the T-S fuzzy neural network controller could be better dynamic performance, stable performance, anti-disturbance and robustness.Towards to double inverted pendulum, it reduces the input variable dimension of the fuzzy controller by designing a fusion function using optimization control theory. And then, it designs the membership function and the fuzzy rule of Mamdani fuzzy controller. The question of "rule explosion" is solved successfully. The simulation results of fuzzy controller and LQR controller show that the fuzzy controller proposed in this paper could be better dynamic performance, stable performance and robustness.