Learning Techniques and Neural Networks for the Solution of N-Stage Nonlinear Nonquadratic Optimal Control Problems
This paper deals with the problem of designing closed-loop feed-forward control strategies to drive the state of a dynamic system (in general, nonlinear) so as to track any desired trajectory joining the points of given compact sets, while minimizing a certain cost function (in general, nonquadratic). Due to the generality of the problem, conventional methods (e.g., dynamic programming, maximum principle, etc.) are difficult to apply. Then, an approximate solution is sought by constraining control strategies to take on the structure of multi-layer feed-forward neural networks. After discussing the approximation properties of neural control strategies, a particular neural architecture is presented, which is based on what has been called the “Linear-Structure Preserving Principle” (the LISP principle). The original functional problem is then reduced to a nonlinear programming one, and backpropagation is applied to derive the optimal values of the synaptic weights. Recursive equations to compute the gradient components are presented, which generalize the classical adjoint system equations of N-stage optimal control theory. Simulation results related to non-LQ problems show the effectiveness of the proposed method.
KeywordsCost Function Optimal Control Problem Synaptic Weight State Trajectory Optimal Control Strategy
Unable to display preview. Download preview PDF.
- A. P. Sage. Optimum Systems Control. Prentice-Hall, Englewood Cliffs, 1968.Google Scholar
- A. E. Bryson and Y. C. Ho. Applied Optimal Control. Blaisdell Publishing Company, 1969.Google Scholar
- R. E. Larson. State Increment Dynamic Programming. American Elsevier Publishing Company, 1968.Google Scholar
- D.L. Kleinman and M. Athans. “The Design of Suboptimal Linear Time-Varying Systems”. IEEE Trans. Automatic Control, Vol. AC-13, pp. 150–159, 1968.Google Scholar
- P. M. Mäkilä and H. T. Toivonen. “Computational Methods for Parametric LQ Problems- A survey”. IEEE Trans. Automatic Control, Vol. AC-32, pp. 658–671, 1987.Google Scholar
- T. Parisini and R. Zoppoli. “Neural Networks for the Solution of N-Stage Optimal Control Problems”, Artificial Neural Networks, T. Kohonen, K. Mäkisara, O. Simula, and J. Kangas, Eds., North-Holland, 1991.Google Scholar
- T. Parisini and R. Zoppoli. “Multi-Layer Neural Networks for the Optimal Control of Nonlinear Dynamic Systems”, Proc. First IFAC Symposium on Design Methods of Control Systems, Zurich, Switzerland, 1991.Google Scholar
- A. R. Barron. “Universal Approximation Bounds for Superpositions of a Sigmoidal Function”. Technical Report No. 58, Department of Statistics, University of Illinois, 1991.Google Scholar
- T. Parisini and R. Zoppoli. “Backpropagation and Dynamic Programming in N-Stage Optimal Control Problems”. DIST Int. Rep. No. 90/4, University of Genoa, Italy, 1991.Google Scholar
- Ya. Z. Tsypkin. Adaptation and Learning in Automatic Systems. Academic Press, 1971.Google Scholar
- D. E. Rumelhart and J. L. McClelland. Parallel Distributed Processing, Vol. 1, Chapter 8, Cambridge, MA, MIT Press, 1986.Google Scholar
- P. J. Werbos. “Beyond Regression: New Tools for Prediction and Analysis in the Behavioural Sciences”. Ph.D. Dissertation, Harvard University, Cambridge, MA, 1974.Google Scholar