ABSTRACT
This chapter introduces a family of neural network (NN) control architectures known as adaptive critic controllers as a natural extension of simpler architectures. It provides the readers with basic motivation, background, and description of different adaptive critic controllers by presenting a series of NN adaptive controller architectures ranging from a single NN adaptive controller to Globalized dual heuristic programming. Due to the continuous interaction between the controller and the plant, the quality of a certain control strategy can only be fully measured after analyzing all future effects it has on the control mission, in our case trajectory tracking. The same principle is applied to components and actuators, though it is possible in those cases that more than one output from different elements operating at only a fraction of its total capability is used at the same time. In both examples, faults are simulated by instantly or gradually changing the model of the plant.
