Enhanced robustness of robot manipulators using fixed-time sliding mode control and RBF neural network

This paper presents a control framework that integrates fixed-time sliding mode control (FxTSMC) with a radial basis function neural network (RBF-NN). The novelty of the proposed approach lies in the synergy between the two techniques, leveraging their respective strengths to enhance overall system performance. The FxTSMC component guarantees fast system response within a predefined time, while the RBF-NN offers adaptive capabilities to handle model uncertainties, thereby improving robustness. The stability of the combined approach is theoretically established and validated through its application to a nonlinear system, specifically a 2-DoF robotic manipulator. Comparative simulations with existing methods are conducted, accompanied by discussions on the results.