Wavelet neural network-based H^∞ trajectory tracking for robot manipulators using fast terminal sliding mode control

This paper focuses on fast terminal sliding mode control (FTSMC) of robot manipulators using wavelet neural networks (WNN) with guaranteed H^∞ tracking performance. The FTSMC for trajectory tracking is employed to drive the tracking error of the system to converge to an equilibrium point in finite time. The tracking error arrives at the sliding surface in finite time and then converges to zero in finite time along the sliding surface. To deal with the case of uncertain and unknown robot dynamics, a WNN is proposed to fully compensate the robot dynamics. The online tuning algorithms for the WNN parameters are derived using Lyapunov approach. To attenuate the effect of approximation errors to a prescribed level, H^∞ tracking performance is proposed. It is shown that the proposed WNN is able to learn the system dynamics with guaranteed H^∞ tracking performance and finite time convergence for trajectory tracking. Finally, the simulation results are performed on a 3D-Microbot manipulator to show the effectiveness of the controller.