A model following controller is proposed for discrete nonlinear systems using a recursive multilayer perceptron (MLP). The MLP network contains dynamics and is able to minimize the error between the plant and a desired model in many cases. An example is given.
Proceedings of the 29th IEEE Conference on Decision and Control
Backpropagation algorithms, Control systems, Multilayer perceptrons, Neurofeedback
Abdallah, Chaouki T.; Don Hush; and B. Horne. "Model following using multilayer perceptrons." Proceedings of the 29th IEEE Conference on Decision and Control (1990): 1730-1731. doi:10.1109/CDC.1990.203916.