Journal of South China University of Technology (Natural Science Edition) ›› 2005, Vol. 33 ›› Issue (2): 43-47.

Previous Articles     Next Articles

Control of the Double·Link Inverted Pendulum by Using Takagi-Sugeno Model Based on Gradual Fuzzy Neural Network

Zhu Xue-feng  Zhou Wen-bin  Chen Hua-yan   

  1. College of Automation Science&Engineering,South China Univ.of Tech.,Guangzhou 510640,Guangdong,China
  • Received:2004-05-14 Online:2005-02-25 Published:2005-02-25
  • Contact: 朱学峰(1940-)，男，教授，博士生导师，主要从事工业过程的建模与最优控制、预测控制、人工神经元网络及其在过程控制中的应用、软测量技术与智能控制等方面的研究 E-mail:xfzhu@seut.edu.cn
  • About author:朱学峰(1940-)，男，教授，博士生导师，主要从事工业过程的建模与最优控制、预测控制、人工神经元网络及其在过程控制中的应用、软测量技术与智能控制等方面的研究

PDF

801

Abstract:

A new control method is proposed by using the gradual fuzzy neural network.In this method,the Taka-gi-Sugeno fuzzy neural network controller and gradual fuzzy control rules are adopted to stabilize the double-link in-verted pendulum. Moreover,Genetic Algorithm is used to optimize the parameters of the Takagi-Sugeno fuzzy con-troller in 4 steps.Experimental results indicate that,only with 20 rules，the gradual fuzzy neural network controller
can stabilize the double-link inverted pendulum with good effect and strong system robustness.

Key words: gradual control, inverted pendulum, Takagi-Sugeno fuzzy logic, neural network control