Direct adaptive regulation using recurrent neural networks: modeling error-external disturbance effects
George A. Rovithakis, M.A. Christodoulou · 2002
A robust direct nonlinear adaptive state regulator, for unknown plants that are modeled by recurrent neural networks is discussed. The unavoidable appearance of a modeling error which is not a priori bounded, as well as the effects of both additive and multiplicative external disturbances on the closed loop system are examined. Generally, under certain modifications on the control and update laws, uniform boundedness of all signals in the closed loop is ensured.>