Robust adaptive neural trajectory tracking control of unmanned surface vessels under input saturation

Abstract

In this paper, we investigate the trajectory tracking control problem for unmanned surface vessels (UVSs) with the dynamic uncertainties and unknown external disturbances under input saturation. The input saturation nonlinearity is approximated by a Gaussian error function, the FTNTD is used to remove the derivation operation of virtual control law, and the adaptive NN is applied to reconstruct the dynamic uncertainties of USVs. A novel nonlinear function featured by “large-error versus small-gain, small-error versus large-gain” is embedded into virtual and actual control laws to improve the control performance and relieve the adverse effect of input saturation. Based on these methods, a novel robust adaptive neural trajectory tracking control scheme is proposed using the vector-backstepping design method. By means of a newly constructed non-quadratic Lyapunov function, it is theoretically shown that all the signals in the closed-loop trajectory tracking control system of UVSs are bounded. Finally, simulation results verify the effectiveness of our novel robust adaptive neural trajectory tracking control scheme.