Command Filtered Backstepping Control of a Two-Link Flexible Joint Manipulator with Uncertainties Based on Reduced-Order ESO

The motion control of flexible joint manipulators (FJMs) is a hot topic in the field of robot control, and the uncertainties including parameter perturbations and disturbances are the key challenges in the development of control strategies for FJMs. In this paper, we take a two-link FJM with complex uncertainties as the object, and present a command filtered backstepping control approach based on reduced-order extended state observer (RESO) for the trajectory tracking control of the two-link FJM with high precision. To enhance the robustness of the control system, an RESO which takes parameter perturbations, friction term and external disturbances as the lumped disturbances is constructed to estimate and compensate for the disturbances. The second-order command filter technique is introduced to eliminate the "explosion of complexity" problem of the conventional backstepping method, and an error compensation dynamic system is developed to reduce the potential filtering errors. All signals of the closed-loop system are proved to be uniformly ultimately bounded by Lyapunov theorem. Simulation reaults are shown to demonstrate the effectiveness and efficiency of the proposed control method.