Adaptive FIT-SMC Approach for an Anthropomorphic Manipulator With Robust Exact Differentiator and Neural Network-Based Friction Compensation

In robotic manipulators, feedback control of nonlinear systems with fast finite-time convergence is desirable. However, because the parametric and model uncertainties, robust tuning manipulators pose many challenges related to trajectory tracking system. This research proposes a state-of-the-art algorithm, which combination integral terminal sliding mode (FIT-SMC), exact differentiator (RED) observer, feedforward neural network (FFNN) based estimator. Firstly, dynamic manipulator established for n-degrees freedom (DoFs) system by taking into account LuGre friction model. Then, FIT-SMC compensation-based has been proposed manipulator. addition, RED observer developed get estimates joints’ velocities. Since state unmeasurable, FFNN training estimating torque. The Lyapunov method presented demonstrate enforcement approach simulated in MATLAB/Simulink environment compared no characterize performance. Simulation results obtained strategy affirm its effectiveness multi-DoF model-based compensation having an overshoot settling time less than 1.5% 0.2950 seconds, respectively, all joints