Computer Algebra as a Tool for Solving Optimal Control Problems

  • C. Gomez
  • J. P. Quadrat
  • A. Sulem


Solving optimal control problems usually consists in modelizing type problem, proving the existence of a solution, choosing a method leadint to a numerical solution and writing a program (generally in FORTRAN). These steps can be automatized using formal calculus and inference techniques. An expert system in stochastic control is presented. It is embedded in the MACSYMA system. After the description of the problem by the user in semi-natural language, the system generates the equations which modelize the problem. Then by using PROLOG (for encoding the used theorems) and symbolic manipulations on the equations, the system can prove the existence of a solution. Finally the system chooses a numerical method to solve it using symbolic differentiations and mtrix inversions and it generates the associated FORTRAN program.


Expert System Optimal Control Problem Computer Algebra Stochastic Control Symbolic Manipulation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    MACSYMA Reference Manual. Version 10. The Mathlab Group, MIT, Laboratory for Computer Science. MIT. December 1983.Google Scholar
  2. 2.
    Gloess P., Logis User’s Manual. Second Edition. UTC/GI, BP 233, 60206 Compiègne Cedex, France. June 1984.Google Scholar
  3. 3.
    Davis J. and Moon D., Multics Maclisp Reference Manual, Revision 1, MIT, Laboratory for Computer Science, 1980.Google Scholar
  4. 4.
    Pitman K.M., The revised MACLISP manual, MIT, Laboratory for Computer Science, 1983.Google Scholar
  5. 5.
    Queinnec C., LISP: Langage d’un autre type, Edition Eyrolles, 1983.Google Scholar
  6. 6.
    Adams R.A., Sobolev spaces. Academic Press. 1975.MATHGoogle Scholar
  7. 7.
    Lions J.L. and Magenes E., Problèmes aux limites non homogènes et applications, Tome I, Dunod, Paris, 1968.Google Scholar
  8. 8.
    Lions J.L., Quelques méthodes de résolution des problèmes aux limites non linéaires. Dunod, Gauthier-Villars, Paris, 1969.Google Scholar
  9. 9.
    Charniak E., MacDermott D. and Riesbeck C., Artificial Intelligence Programming. Lawrence Erlbaum Associates, Hillsdale, New Jersey. 1979.Google Scholar
  10. 10.
    Bensoussan A. et al., Commande optimale de systèmes stochastiques. RAIRO, Systems Analysis and Control, Vol. 18, n°2. 1984.Google Scholar
  11. 11.
    Blankenship G.L. et al., An expert system for control and signal processing with automatic FORTRAN code generation. CDC 1984, Las Vegas, to appear.Google Scholar
  12. 12.
    Gomez C., Quadrat J.P. and Sulem A., Towards an expert system in stochastic control: the Hamilton-Jacobi equation part. Nice, Juin 1984. Springer Verlag, Lecture Notes in Control and Information Sciences, 62.Google Scholar
  13. 13.
    Gomez C., Quadrat J.P. and Sulem A., Towards an expert system in stochastic control: optimization in the class of local feedbacks. Rome, 1984. To appear.Google Scholar
  14. 14.
    Gloess P., Understanding Expert systems. Alfred Handy Guide. 1984. To appear.Google Scholar
  15. 15.
    Feigenbaum E.A. and Barr A., The Handbook of Artificial Intelligence, Heuristics Press, Stanford, California, W. Kaufmann Inc., Los Altos, California. 1982.Google Scholar
  16. 16.
    Gloess P. and Nguyen D., OBLOGIS, an object oriented implementation of PROLOG in a LISP environment. Colloque International d’Intelligence Artificielle, Marseille, France, October 1984.Google Scholar

Copyright information

© Kluwer Academic Publishers 1985

Authors and Affiliations

  • C. Gomez
    • 1
  • J. P. Quadrat
    • 1
  • A. Sulem
    • 1
  1. 1.Domaine de VoluceauInstitut National de Recherche en Informatique et en, Automatique(INRIA)Le Chesnay CedexFrance

Personalised recommendations