Super-Twisting Algorithm Sliding Mode Control of Flexible Manipulators Considering Uncertainty

The dynamic model of a spatial flexible manipulator is challenging to establish accurately and is susceptible to disturbances from uncertainties, which negatively impact system control. To mitigate the effects of uncertainties and improve control precision, this paper develops the dynamic model of a spatial flexible manipulator system using the Lagrange principle and the assumed mode method. Additionally, to address the chattering issue in sliding mode control, a super-twisting algorithm sliding mode control strategy is proposed, which utilizes an RBF neural network to compensate for system uncertainties. By incorporating the concept of higher-order sliding mode, the super-twisting algorithm sliding mode is integrated with non-singular terminal sliding mode to design the sliding surface and control law, ensuring the stable and rapid convergence of the system state while reducing chattering. The simulation results demonstrate that the proposed strategy reduces fluctuations in rotation angle and angular velocity errors at both the input and output ends, enhances trajectory tracking performance, and achieves effective chattering suppression.