Adaptive Neural Sliding Mode Based Control of Robot Manipulators with Disturbance Observer

For robotic manipulator systems experiencing uncertainties and external disturbances, an integral nonsingular fast terminal sliding mode control (INFTSMC) approach is introduced, incorporating an adaptive sliding mode disturbance observer (ASMDO). To relax the constraints on the design of the robotic arm disturbance observer, an adaptive method is added to estimate the upper bound of the disturbances. The fixed-time stability of the proposed ASMDO is analyzed theoretically. Secondly, radial basis function neural network (RBFNN) is utilized to estimate the system uncertainty. Subsequently, by combining the proposed observer with the RBFNN, the INFTSMC is structured around an integral nonsingular fast terminal sliding surface. The stability and finite-time convergence of the full system are then demonstrated through Lyapunov theory, and simulation outcomes confirm the proposed method's effectiveness.