Neuro-controller design for nonlinear fighter aircraft maneuver using fully tuned RBF networks (Q5943847)

A stable on-line control scheme is proposed for a nonlinear aircraft controller design. This scheme consists of a conventional controller and a neural controller, which includes a growing radial basis function network (GRBFN). A Lyapunov stability theory approach is used to derive the tuning rule for updating all the parameters of the GRBFN (weights, widths and centers of the Gaussian functions). In the simulation study, the developed control strategy and the proposed tuning rule are applied to a high performance fighter aircraft, which executes a high \(\alpha\) (angle of attack) stability-axis roll maneuver.