Adaptive Model-free Non-singular Terminal Sliding-mode Control Approach for Fast Convergence Rate in Robot Manipulators

This paper presents a new model-free non-singular terminal sliding-mode control, then applies the control approach to robot manipulator. The proposed control approach uses one-sample delayed information to cancel out system uncertainties and disturbances. The proposed one offers a new pole-placement method with non-singular terminal sliding variable and hence aims to achieve strong attraction without singularity within a finite time. Time-varying switching gain is employed to always guarantee system stability, which is designed to be requiring no knowledge of stability criteria. The proposed control approach was demonstrated to be uniformly ultimately bounded, achieving Lyapunov stability. The effectiveness of the proposed control approach is compared to that of the existing control approach.