Nonlinear adaptive control using gaussian networks with composite adaptation for improved convergence

Abstract

The use of composite adaptive laws for control of the affine class of nonlinear systems having unknown dynamics is proposed. These dynamics are approximated by Gaussian radial basis function neural networks whose parameters are updated by a composite law that is driven by both tracking and estimation errors, combining techniques used in direct and indirect adaptive control. This is motivated by the need to improve the speed of convergence of the unknown parameters, hence resulting in a better system performance. The inherent approximation error of the neural networks might lead to instability because of parameter drift. This is compensated for by augmenting the control law with a low gain sliding mode component and using deadzone adaptation for the indirect part of the composite law. The stability of the system is analysed and the effectiveness of the method is demonstrated by simulation and comparison with a direct adaptive control scheme.