Neural Networks-Based Fixed-Time Control for a Robot with Uncertainties and Input Deadzone

In this paper, neural networks-based fixed-time control is presented for a robot with uncertainties and actuator input deadzone. Model-based fixed-time control for the robot has been proposed based on an assumption parameters of the robot are completely known. To deal with the uncertainties of the robot and improve the stability of the robot, the adaptive neural network approximation method is proposed. Compared with existing works, the proposed control scheme can not only eliminate the influence of input deadzone, but also make all the state variables converge to a small region of zero in fixed-time, which can accelerate the convergence speed and improve the transient performance of the tracking control for the robot. In order to verify the feasibility of the proposed control algorithm, extensive simulations are carried on two joint robot manipulator.