A bioinspired filtered backstepping based approach to tracking control of underactuated hovercraft

This paper studies the trajectory tracking control problem of an underacutated hovercraft, equipping only two control inputs with three degrees of freedom. A novel bioinspired tracking control approach is proposed by intergrating backstepping technique and additive neural dynamics model. The controller is first designed from the error dynamics analysis of the hovercraft. And then three additive neural dynamics derived from biological membrane equation are employed to filter the virtual control signals, avoiding the inherent complexity of numerical derivatives in the backstepping. The proposed tracking controller is thus computationally efficient, and capable of driving both the position tracking error and the orientation error to converge to a small neighborhood of the origin, achieving a practical stability. The effectiveness of the proposed control approach is demonstrated by simulations.