ANN-Based vibration control of an aerial refueling hose system with input nonlinearity and prescribed output constraint

Abstract

An adaptive artificial neural network (ANN) boundary control scheme is developed for an autonomous aerial refueling hose system in this study. The objectives of this paper are 1) to guarantee the state of the system uniformly bounded, 2) to make the system output meet the prescribed performance, that is, −χa(t)<y(l(t),t)<χb(t). The proposed novel control law can achieve control objectives by suppressing vibrations of the hose. The unmodeled dynamics of the system are handled by the approximation ability of the designed ANN. The Nussbaum function is employed to construct an auxiliary control signal to deal with the input saturation nonlinearity of the system. By applying the novel control law, the state of the closed-loop system is uniformly bounded which is proved by Lyapunov’s direct method. Without violating the asymmetric time-varying prescribed function on system output is obtained by the barrier Lyapunov function. Extensive numerical simulations are performed to confirm the efficiency and efficacy of the proposed novel control strategy which allows multiple control objectives. The simulation results show that the output constraint is never violated, which is guaranteed by the asymmetric Lyapunov function.