RNN-Based cooperative motion control of 2-dof Robot Arms

Yingda Dai, Masami Konishi, Jun Imai

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

This paper presents a general recurrent neural network (RNN) model for online control of time-varying robot manipulators. The robot manipulators with different setting parameters work cooperatively on an unknown curve tracing. Each joint of the manipulator is respectively provided a learning method to optimize trajectory by the training RNN model. In this paper, the proposed RNN model shortens the period of learning and improves the cooperative accuracy of the existing neural networks for solving problems such as cutting or welding special types of products. A More complicated construction is to fit it for online cooperation. Simulation results show the effectiveness of this approach and that the proposed RNN model can successfully learn the inverse dynamics of robot manipulators as well as perform accurate tracking for a general trajectory. It is also shown that the proposed method is better than the conventional method due to its improved evolution functions.

Original languageEnglish
Pages (from-to)937-952
Number of pages16
JournalInternational Journal of Innovative Computing, Information and Control
Volume3
Issue number4
Publication statusPublished - Aug 2007

    Fingerprint

Keywords

  • Collision avoidance
  • Cooperative motion control
  • Kobot arm
  • Recurrent neural network (KNN)
  • Trajectory generation

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this