RBF neural network based dynamic surface control for anti-skid braking system using barrier Lyapunov theorem

Abstract

A method of dynamic surface control, which is based on RBF neural network, is proposed for aircraft anti-skid braking system (ABS). The proposed scheme solves a difficult problem of controller design in the present of output constraints and unknown external disturbance. In this approach, using the RBF neural network approximates to the uncertain nonlinearities of ABS. Subsequently, we demonstrate that the proposed controller can guarantee the boundedness of the output constraints by developing the asymmetric barrier Lyapunov function (ABLF), which makes the wheel slip-ratio constraints more flexible for various runway surfaces and runway transitions. The back-stepping strategy based on dynamic surface control (DSC) is introduced to eliminate repeated differentiation resulting from ABLF synthesis. The proposed scheme can guarantee the stability of the overall system with unknown external disturbance. The results of simulations have validated the effectiveness of the proposed control scheme.