Subscribe Now Subscribe Today
Science Alert
 
FOLLOW US:     Facebook     Twitter
Blue
   
Curve Top
Journal of Applied Sciences
  Year: 2008 | Volume: 8 | Issue: 21 | Page No.: 3783-3794
DOI: 10.3923/jas.2008.3783.3794
Nonlinear Generalized Predictive Controller Based on Artificial Neural Network for Robot Control
B. Durmus and N. Yumusak

Abstract:
This study deals with the tracking control problem of a robotic manipulator with changing dynamics. A multiple-input multiple-output (MIMO) artificial neural network based generalized predictive control (NGPC) controller was designed for a six-degrees-of-freedom (6-DOF) robotic manipulator random disturbances and changing load. A three-layered neural network was used in the controller design to predict robotic manipulator inputs which track a desired trajectory. Standard back propagation (BP) algorithm was used as a learning algorithm to minimize the difference between actual trajectory and that predicted by the neural network. NGPC controller was compared the conventional GPC under different control conditions. Results show that proposed control improved the ability of GPC under uncertainties.
PDF Fulltext XML References Citation Report Citation
 RELATED ARTICLES:
  •    Intelligent Humidity Control for Healthy Home to Wealthy Industry: A Review
  •    Development of Artificial Neural Network for Optimisation of Reusability in Automotive Components
How to cite this article:

B. Durmus and N. Yumusak, 2008. Nonlinear Generalized Predictive Controller Based on Artificial Neural Network for Robot Control. Journal of Applied Sciences, 8: 3783-3794.

DOI: 10.3923/jas.2008.3783.3794

URL: https://scialert.net/abstract/?doi=jas.2008.3783.3794

 
COMMENT ON THIS PAPER
 
 
 

 

 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 

       

       

Curve Bottom