On adaptive critic architectures in feedback control

Two feedback control systems are designed that employ the adaptive critic architecture, which consists of two neural networks, one of which (the critic) tunes the other. The first application is a deadzone compensator, where it is shown that the adaptive critic structure is a natural consequence of the mathematical problem of inversion of an unknown function. In this situation the adaptive critic appears in the feedforward loop. The second application is the supervisory loop adaptive critic, where it is shown that the critic neural network requires additional dynamics that effectively give it a memory capability