Abstract
A novel controller using second-order fast nonsingular terminal sliding mode control (SOFNTSMC) and adaptive neural networks (ANNs) is proposed for rehabilitation robots to achieve tracking control. First, the mathematical model is built as a two-link single-arm robot with uncertainties. The SOFNTSMC is employed to achieve a faster convergence, singularity-avoidance and reduce chattering with high precision. Then, ANNs are introduced to handle the model uncertainties and disturbances without any prior knowledge. Subsequently, using the finite-time stability analysis, all signals are bounded in the closed-loop system. Finally, comparative simulations are performed on a two-joints single-arm robot to verify the effectiveness of the proposed controller.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have