A Lyapunov-based design of robust control for a robot using neural networks

Abstract

An adaptive neural control scheme for mechanical manipulators is presented. The design basically consists of a neural controller which implements a feedback linearization control law for a generic manipulator with unknown parameters, and a sliding-mode control which robustifies the design and compensates for the neural approximation errors. The neural updating law is obtained based on the Liapunov design to guarantees the closed-loop stability. Moreover, this scheme attains a good trajectory tracking with a small transient under the robot dynamical uncertainties.