Position tracking of an autonomous underwater glider based on adaptive filtered backstepping control.

This paper considers the problem of position tracking control of the autonomous underwater gliders, which operates under the environment of model uncertainties and external disturbances. The motion of the glider is controlled by ballast mass and moving mass, which have limited dynamic range. An adaptive control system for position tracking based on filtered backstepping is proposed, and neural networks and sliding mode control are used to improve robustness. Next, using Lyapunov stability theory, it is proved that all the signals are bounded and tracking errors eventually converge to a neighborhood of zero. The presented controller can be easily put into practice without the requirement to know the parameters of the glider and the value of external environmental disturbances. Finally, the effectiveness and good robustness of the proposed control system are verified by simulation results. • The position is taken as the control target for the six degrees of freedom movement of AUG. • The filtered backstepping method is used to solve the problem of "explosion of complexity". • The neural networks are used to solve the problem of model uncertainties. • The SMC is used to solve the problem of the disturbances and the approximation errors of neural networks. [ABSTRACT FROM AUTHOR]