Skip to Main Content
A new model reference adaptive control design method using neural networks that improves both transient and steady-state performance is proposed in this paper. Stable tracking of a desired trajectory can be achieved for nonlinear systems having significant uncertainties. An uncertainty-state observer structure is designed to achieve desired transient performance. The neural network adaptation rule is derived using Lyapunov theory, which guarantees stability of the error dynamics and boundedness of the neural network weights. An extra term is added in the controller expression to introduce a “soft-switching” sliding mode that can be used to reduce tracking error. The proposed design method is applied to control the velocity and position of an electrohydraulic piston comprising industrial components and having a limited bandwidth, and experimental results demonstrate its effectiveness as compared to commonly used controllers.
Date of Publication: June 2013