Wavelet Adaptive Control for Robotic Manipulator With Input Deadzone

Abstract

Abstract In this paper, the adaptive wavelet neural network (WNN) tracking control problem is investigated for robotic manipulator with input deadzone. The WNN is used to approximate the unknown nonlinear function and the derivative of virtual control in the system, which avoids the problem “explosion of complexity” in the traditional backsteppinng control methods, the requirement of input instruction signal is reduced. The robust term is designed to compensate the approximation errors of WNN and external disturbance. Since wavelet functions are used in WNN, its learning capability is precede to the traditional neural network for system identification. It is proved that the proposed controller can guarantee that all signals of the closed-loop system are uniformly ultimately bounded. Simulation results demonstrate the effectiveness of the proposed approach.