Neural Network Modeling of a Flexible Manipulator Robot - Archive ouverte HAL Access content directly
Conference Papers Year : 2012

Neural Network Modeling of a Flexible Manipulator Robot

(1) , (1)
Rahma Boucetta
  • Function : Author
  • PersonId : 1004696
Mohamed Naceur Abdelkrim
  • Function : Author
  • PersonId : 865626


This paper presents an artificial neural networks application for a flexible process modeling. A flexible planar single-link manipulator robot is considered. The dynamic behavior of this process is described using Lagrange equations and finite elements method. The artificial neural networks are all variations on the parallel distributed processing (PDP) idea. The architecture of each network is based on very similar building blocks which perform the processing. Therefore, two feed-forward and recurrent neural networks are developed and trained using back-propagation algorithm to identify the dynamics of the flexible process. Simulation results of the system responses are given and discussed in terms of level of error reduction. Finally, a conclusion encloses the paper.
Fichier principal
Vignette du fichier
978-3-642-33260-9_34_Chapter.pdf (254.52 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01551709 , version 1 (30-06-2017)


Attribution - CC BY 4.0



Rahma Boucetta, Mohamed Naceur Abdelkrim. Neural Network Modeling of a Flexible Manipulator Robot. 11th International Conference on Computer Information Systems and Industrial Management (CISIM), Sep 2012, Venice, Italy. pp.395-404, ⟨10.1007/978-3-642-33260-9_34⟩. ⟨hal-01551709⟩
57 View
75 Download



Gmail Facebook Twitter LinkedIn More