Fuzzy Control of Linear Flexible Double Inverted Pendulum System

• Published in: 2012 International Conference on Control Engineering and Communication Technology pp. 342 - 345
• Main Authors: Jimin Yu, Linyan Huang, Shangbo Zhou
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.12.2012
• Subjects: Educational institutions; Equations; Flexible Inverted Pendulum; Fuzzy control; Fuzzy neural networks; Input variables; Lagrange equation; LQR; Mathematical model; Neural networks; Sugeno-type fuzzy neural network
• ISBN: 9781467344999; 1467344990
• DOI: 10.1109/ICCECT.2012.148

In this paper, Lagrange equation is used to derive the mathematical model of linear double flexible inverted pendulum system, which simplifies the modeling process. As the flexible inverted pendulum system is a nonlinear, multivariable, strong coupling, and unstable control system. In order to improve the double flexible real-time control of inverted pendulum system response speed and stability, a LQR controller which can stabilize the inverted pendulum system is designed, according to this, an more efficient neural network controller is designed which is based on the Sugeno-type fuzzy inference rules. The controller takes the hybrid of BP neural network and least squares algorithm to train parameters, which can accurately summarize the amount of input and output fuzzy membership functions and fuzzy logic rules. By comparing the simulations, it proves that Sugeno-type fuzzy neural network controller is better than LQR controller in stability, speed and control accuracy.