In this paper, the applications of neural networks for the design of learning controllers are discussed. It is argued that the usual error back propagation (EBP) algorithm cannot be readily used for the training of neural controllers. Instead, in order to ensure the convergence of the training process and the stability of the closed-loop system, a stability approach must be taken to derive a learning algorithm. We use Liapunov's stability approach to develop a learning rule for neural network controllers that would guarantee the stability of the training process under mild conditions, These controllers do not require a priori information about the plant dynamics. The designed controller is then used for the control of robots.
展开▼
