Method of Controlling Inverted Pendulum Based on Fusing Genetic Algorithm and Neural Network

Abstract:

Article Preview

As a result of inverted pendulum control based on traditional neural network algorithm existing shortcomings that are training time to be long, the convergence rate slow and easy to fall into the partial minimum point, a method of fusing genetic algorithm and neural network control is proposed in this article. The structure of inverted pendulum controller adopts neural network, genetic algorithm is used to optimize the attached weights and thresholds of neural network. The experimental results illustrate that the method can overcome inadequacies of the neural network, with the characteristics of convergence speed is fast, global search ability is strong, dynamic and steady state control performance is good. Furthermore, it has advantages of simple controller structure and easy to be implemented.

Info:

Periodical:

Advanced Materials Research (Volumes 383-390)

Edited by:

Wu Fan

Pages:

5758-5763

DOI:

10.4028/www.scientific.net/AMR.383-390.5758

Citation:

L. N. Liu and Y. Wang, "Method of Controlling Inverted Pendulum Based on Fusing Genetic Algorithm and Neural Network", Advanced Materials Research, Vols. 383-390, pp. 5758-5763, 2012

Online since:

November 2011

Authors:

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.