Journal of Intelligent and Robotic Systems

, Volume 15, Issue 1, pp 3–10 | Cite as

A neural network based adaptive robot controller

  • A. S. Morris
  • S. Khemaissia
Article

Abstract

Neural network based adaptive controllers have been shown to achieve much improved accuracy compared with traditional adaptive controllers when applied to trajectory tracking in robot manipulators. This paper describes a new Recursive Prediction Error technique for estimating network parameters which is more computationally efficient. Results show that this neural controller suppresses disturbances accurately and achieves very small errors between commanded and actual trajectories.

Key words

Neural networks adaptive robot control recursive prediction error 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Azimi-Sadjadi, M., et al.: Supervised learning process of multi-layer perceptron neural networks using fast least squares, in Proc. IEE Int. Conf. on Acoustics, Speech and Signal Processing, 1990, pp. 1381–1384.Google Scholar
  2. 2.
    Billings, S. A. et al.: A comparison of the back propagation and recursive prediction error algorithms for training neural networks, Mechanical Systems and Signal Processing 5 (1991), 233–255.Google Scholar
  3. 3.
    Chen, S., et al.: Non-linear systems identification using neural networks, Int. J. Control 51 (1990), 1191–1214.Google Scholar
  4. 4.
    Kawato, M., et al.: Hierarchical neural network model for voluntary movement with application to robotics, IEE Control Systems Magazine (1988), 8–16.Google Scholar
  5. 5.
    Khemaissia, S. and Morris, A. S.: Neuro-adaptive control of robotic manipulators, Robotica 11 (1993), 465–473.Google Scholar
  6. 6.
    Ljung, L. and Soderstrom, T.: Theory and Practice of Recursive Identification, MIT Press, Cambridge, MA, 1983.Google Scholar
  7. 7.
    Tomochika, O., et al.: Trajectory control of robotic manipulators using neural networks, IEEE Trans. Industrial Electronics 38 (1991), 195–202.Google Scholar
  8. 8.
    Sun, J., et al.: A fast algorithm for finding global minima of error functions in layered neural networks, Proc. IEEE Int. Joint Conf. on Neural Networks, 1 (1990), 715–720.Google Scholar

Copyright information

© Kluwer Academic Publishers 1996

Authors and Affiliations

  • A. S. Morris
    • 1
  • S. Khemaissia
    • 1
  1. 1.Robotics Research Group, Department of Automatic Control and Systems EngineeringsUniversity of SheffieldSheffieldU.K.

Personalised recommendations