Abstract
Numerous paper on CMAC and adaptive control, have been published. However few of them relate effective hardware real-time implementation. G.Kraft [Kraft 90] exposes an adaptive scheme, based on a first order processes (i.e. integrator). This application brings interesting simulation results, but first order process is not useful with real processes.
First we describe briefly a modified CMAC paradigm, with an on-line computation of connections to reduce memory size and allowing real-time process control. In a second step, we present the DSP architecture car, we have developed, and give comparisons results in learning and transfer phase between different processors.
Then in a third phase, we expose the modified Kraft structure, with a second order unknown process and we detail the result of a manipulation and give experimental results.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
Bibliography
Hornik K. Multilayer feedforward networks ar e universal approximators, Neural Networks, 2, pp 359–366-1989.
Jordan MI. Generic constraints on underspecified target trajectories. Proceedings of International Joint Conference On neural Networks. Washington Vol I, pp 217–225.
Kraft G & Campagna D.A Comparison between CMAC Neural Network Control and two traditionnal Adaptive Control Systems. American Control Conference Pittsburg 1989 pp 21–23.
G.Mercier,K. Madani, C.Barret Various Neural Network Implementations on a low cost and versatile DSP card, EUFIT 94 pp1482–1488.
G.Mercier, K Madani A new on-line CMAC Algorithm for Real tTme Applications, R.I Informatics congress St petersbourg, May 94.
Narandra KS & Partharsaraty K. Stable Adaptative Systems, Englewood Cliffs,NJ Prentice Hall 1989.
Narandra KS & Partharsaraty K. Identification and control of dynamic systems using neural networks. IEEE Transactions on Neural Networks, Vol 1 pp 4–27.
Psaltis,. D., Sideris A., &Yamamura A., Neural controllers,Proceedings of IEEE first International Conference on Neural Networks, San Diego, Vol 4, pp 551–558.
Saerens M.&Soquet. M, A neural controller, Proceedings of IEEE first International Conference on Neural Networks,London pp 211–215.
Saerens M.&Soquet. M, Neural Controller based on Backpropagation Algorithm.IEEE peoceeding F-138, pp 55–62.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mercier, G., Madani, K. (1995). CMAC real-time adaptive control implementation on a D.S.P. based card. In: Mira, J., Sandoval, F. (eds) From Natural to Artificial Neural Computation. IWANN 1995. Lecture Notes in Computer Science, vol 930. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-59497-3_292
Download citation
DOI: https://doi.org/10.1007/3-540-59497-3_292
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-59497-0
Online ISBN: 978-3-540-49288-7
eBook Packages: Springer Book Archive