Coevolutionary Process Control

  • J. Paredis


This text describes the use of a coevolutionary genetic algorithm (CGA) for process control. A CGA combines two artificial life techniques - life-time fitness evaluation (LTFE) and coevolution - to improve the genetic search for a neural network (NN) controlling a given process.

Here, the approach is illustrated and tested on a well-known bioreactor control problem which involves issues of delay, nonlinearity and instability.


Constraint Satisfaction Input Node Basic Cycle Noise Pattern Genetic Search 
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]
    C.W. Anderson and W.T. Miller. Challenging control problems. In T. Miller W, R.S. Sutton, and P. J. Werbos, editors, Neural Networks for Control. MIT Press/Bradford Books, 1991.Google Scholar
  2. [2]
    W.D. Hillis. Co-evolving parasites improve simulated evolution as an optimization procedure. In C.G. Langton, C. Taylor, J. D. Farmer, and S. Rasmussen, editors, Artifical Life II. Addison-Wesley, California, 1992.Google Scholar
  3. [3]
    K.S. Narendra and K. Parthasarathy. Identification and control of dynamical systems using neural networks. IEEE Transaction of Neural Networks. 1(1), 1990.Google Scholar
  4. [4]
    J. Paredis. The evolution of behaviour: Some experiments. In Meyer and Wilson, editors, From Animals to Animats. MIT Press/Bradford Books, 1991.Google Scholar
  5. [5]
    J. Paredis. Coevolutionary constraint satisfaction. In Y. Davidor, H-P. Schwefel, and R. Manner, editors, Parallel Problem Solving from Nature III, Lecture Notes in Computer Science, volume 866. Springer-Verlag, 1994.Google Scholar
  6. [6]
    J. Paredis. Steps towards coevolutionary classification neural networks. In R. Brooks and P. Maes, editors, Proc. Artificial Life IV. MIT Press/Bradford Books, 1994.Google Scholar
  7. [7]
    J. Paredis. The symbiotic evolution of solutions and their representations. In L. Eshelman, editor, Proceedings of the Sixth International Conference on Genetic Algorithms. Morgan Kaufmann Publishers, 1995.Google Scholar
  8. [8]
    J. Paredis. Coevolutionary computation. Artificial Life Journal, 2(4), 1996.Google Scholar
  9. [9]
    J. Paredis. Coevolutionary life-time learning. In H-M. Voigt, M. Ebeling, I. Rechenberg, and H-P. Schwefel, editors, Parallel Problem Solving from Nature IV, Lecture Notes in Computer Science, volume 1141. Springer-Verlag, Heidelberg, 1996.Google Scholar
  10. [10]
    J. Paredis. Symbiotic coevolution for epistatic problems. In Proceedings of the European Conference on Artificial Intelligence, Chichester, 1996. John Wiley and Sons.Google Scholar
  11. [11]
    L.H. Ungar. A bioreactor benchmark for adaptive network-based process control. In T. Miller, W.R.S. Sutton, and P.J. Werbos, editors, Neural Networks for Control. MIT Press/Bradford Books, 1991.Google Scholar
  12. [12]
    D. Whitley. Optimizing neural networks using faster, more accurate genetic search. In Proc. Third Int. Conf. on Genetic Algorithms. Morgan Kaufmann, 1989.Google Scholar
  13. [13]
    D. Whitley. Genetic reinforcement learning for neurocontrol problems. Machine Learning, 13:259–284, 1993.CrossRefGoogle Scholar
  14. [14]
    A.P. Wieland. Evolving controls for unstable systems. In D. Touretzky, J.L. Elman, T.J. Sejnowski, and G.E. Hinton, editors, Proc. of the 1990 Summer School. Morgan Kaufmann, 1991.Google Scholar

Copyright information

© Springer-Verlag Wien 1998

Authors and Affiliations

  • J. Paredis
    • 1
  1. 1.RIKS / MATRIKSUniversiteit MaastrichtMaastrichtThe Netherlands

Personalised recommendations