Backstepping-based adaptive control for underactuated systems

Abstract

A backstepping-based adaptive control scheme is presented for uncertain underactuated systems. On the basis of backstepping approach, virtual control components are first constructed and final control term is then derived by defining an integrated error surface. Proposed control scheme utilises wavelet neural network for the approximation of system uncertainties and thus relaxes the constraint of fully known system dynamics. Update laws for wavelet network parameters are derived in Lyapunov sense. Proposed control scheme guarantees the uniform ultimate boundedness of the closed loop adaptive system. Simulation included illustrates the effectiveness of the proposed approach.