Online trained neural networks have become popular in designing robust and adaptive controllers for dynamic systems with uncertainties in their system equations because of their universal function approximation property. The uniqueness of the method proposed in this work is that the online function approximating network can be used to re-optimize in real time an existing Single Network Adaptive Critic (SNAC) based optimal controller that has already been designed for a nominal system. The controller redesign is carried out in two steps: (i) synthesis of an online neural network that captures the unknown functions in the plant equations on-line (ii) re-optimization of the existing optimal SNAC controller to drive the states of the plant to a desired reference by minimizing a predefined cost function. The neural network weight update rule for the online networks has been derived using Lyapunov theory that guarantees both the stability of the error dynamics as well as boundedness of the weights. This approach has been applied to control the frictional dynamics of a nano-scale array of particles to track a predefined value of average sliding velocity. Here, the system is in an output feedback form where all states are not available for measurement. This is a significant step towards optimal control of systems at nano levels. Numerical results from simulation studies are presented.

Volume Subject Area:
Dynamic Systems and Control
Topics:
Control equipment, Artificial neural networks, Dynamics (Mechanics), Design, Dynamic systems, Errors, Feedback, Function approximation, Nanoscale phenomena, Optimal control, Optimization, Particulate matter, Simulation, Stability, Uncertainty, Weight (Mass)
1.
Bryson, A. E. and Ho, Y. C., “Applied Optimal Control,” Taylor and Francis, 1975.
2.
Werbos, P. J., “Approximate Dynamic Programming for Real-time Control and Neural Modeling.” In White D.A., & Sofge D.A (Eds.), Handbook of Intelligent Control, Multiscience Press, 1992.
3.
Prokhorov
D. V.
, and
Wunsch
D. C.
, “
Adaptive Critic Designs
,”
IEEE Transactions on Neural Networks
, Vol.
8
,
1997
, pp.
997
–
1007
.
4.
Padhi, R., “Optimal Control of Distributed Parameter Systems Using Adaptive Critic Neural Networks,” 2001, Ph.D. Dissertation, University of Missouri Rolla.
5.
Lewis, F.L., Jagannathan, S. and Yesilderik, A., “Neural Network Control of Robot Manipulators and Nonlinear Systems,” Taylor and Francis, UK, 1999.
6.
Leitner
J.
,
Calise
A.
and
Prasad
J. V. R.
, “
Analysis of Adaptive Neural Networks for Helicopter Flight Controls
,”
AIAA Journal of Guidance, Control, and Dynamics
, Vol.
20
, No.
5
, sep-oct 1997, pp.
972
–
979
.
7.
Balakrishnan
S. N.
, and
Biega
V.
, “
Adaptive-Critic Based Neural Networks for Aircraft Optimal Control
,”
Journal of Guidance, Control and Dynamics
, Vol.
19
, No.
4
, July–Aug.
1996
, pp.
893
–
898
.
8.
Braiman, Y., Barhen, J. and Protopopescu, V., “Control of Friction at the Nanoscale,” Physical Review Letters, Vol. 90, No. 9, March 2003.
9.
Ioannou, Petros. A. and Sun, Jing., “Robust Adaptive Cntrol,” Prentice Hall, Inc., 1996.
This content is only available via PDF.
Copyright © 2005
 by ASME
You do not currently have access to this content.