Evolutionary learning algorithms for neural adaptive control (Q4344486)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1034218
Language Label Description Also known as
default for all languages
No label defined
    English
    Evolutionary learning algorithms for neural adaptive control
    scientific article; zbMATH DE number 1034218

      Statements

      0 references
      16 July 1997
      0 references
      learning
      0 references
      nonlinear systems
      0 references
      neural networks
      0 references
      genetic algorithms
      0 references
      chaotic systems
      0 references
      attitude control
      0 references
      Evolutionary learning algorithms for neural adaptive control (English)
      0 references
      The book considers the problem of control in nonlinear physical systems with unknown parameters. Instead of the conventional control methods, which do not solve some tasks successfully, neural networks and genetic algorithms are proposed and employed. The book presents both a technique for modeling complex dynamical systems and a general control architecture using the previously mentioned methods. Chapter 2 gives an overview of system theory, including nonlinear and chaotic systems. Traditional methods of control and a method developed for controlling chaos are summarized here. In Chapter 3, the attitude control problem, which serves as an example throughout the book, is described. Attitude control is the process of orienting a satellite in a specified, predetermined direction. The Euler equations are derived and their chaotic behaviour is shown. This chapter is concluded with a review of numerical methods for solving ordinary differential equations. Chapter 4 reviews the basic forms of artificial neural networks and their properties. Namely, the feedforward networks and the back-propagation algorithm are presented. An algorithm for modeling dynamical systems using artificial neural networks is given in Chapter 5 and in Chapter 6 the techniques for developing neural network controllers are studied. Chapter 7 introduces genetic algorithms and genetic programming and their basic properties. These are used for the attitude control and simulation results are given. Chapters 8 and 9 are devoted to a hybrid adaptive control architecture which combines the two approaches described. The book has a very extensive list of references and the number of illustrations helping to understand the theory is impressive.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references