Search

In this investigation, a model free discrete time neural network control is designed for the trajectory tracking of a kind of nonlinear processes. The introduced control has three main characteristics: (1) the tracking error is used instead of the estimation error in the weights learning equations, avoiding the requirement of a good behavior estimation of the unknown elements in the nonlinear model, (2) the projection method is suggested to avoid the overfitting in the control law, and (3) the Lyapunov technique is utilized to assure the uniform stability of the tracking and weights errors. The suggested technique is applied in two nonlinear processes: the inverted-car and Furuta pendulums. [ABSTRACT FROM AUTHOR]

In this correspondence, we propose an efficient estimator of optimal memory (averaging interval) for discrete-time finite impulse response (FIR) filters in state-space. Its crucial property is that only real measurements and the filter output are involved with no reference and noise statistics. Testing by the two-state polynomial model has shown a very good correspondence with predicted values. Even in the worst case of the harmonic model, the estimator demonstrates practical applicability. [ABSTRACT FROM AUTHOR]