Artificial intelligence techniques for designing switched discrete adaptive controllers for linear time invariant plants

This paper develops a representation of multimodel based controllers using artificial intelligence techniques. These techniques will be neural networks and genetic algorithms. Thus, the interpretation of multimodel controllers in an artificial intelligence frame will allow the application of each specific technique to the design of improved multimodel based controllers. The obtained artificial intelligence-based multimodel controllers are compared with classical single model based ones and with standard multimodel controllers. It is shown through simulation examples that a transient response improvement can be achieved using the proposed artificial intelligence based techniques. Furthermore, a method for synthesizing multimodel based neural network controllers from already designed single model based ones is presented, extending the applicability of this kind of techniques to a more general type of controllers including the case of nonlinear plants. Also, some applications of genetic algorithms to multimodel controller design are proposed. In this way, a method to select a controller with improved robustness properties inspired from the genetic mutation operator is presented. The so obtained scheme has revealed to be adequate to use in noisy environments.