Robust Tracking Control for Lower Limb Rehabilitation Exoskeleton Robots with Multi-Performances

The organization of the lower limb exoskeleton robot (LLRER) is a highly nonlinear system and there is key difficulties including the stability issue and the complex control design. This article uses the feedback linearization method to robustly stabilize LLRER that takes the place of nonlinear function approximator, such as the fuzzy control and the neural network control. The current work “firstly” proposes the difficult nonlinear exponential convergence rate formula of LLRER. The effective operating range of presented algorithm is “global”, whereas the Jacobian linearization method is “locally” valid. The proposed approach simultaneously possesses some features including the almost disturbance rejection feature that replaces the singular perturbation approach without the sufficient requirement that original system should have “complete” condition, and the system stability feature without addressing the difficult Hamilton-Jacobin equation for the H-infinity method. To rigorously verify the effectiveness of the proposed approach, this study uses the AI software “Python” to build a controller-design human–machine interaction package of general nonlinear system. Finally, a comparative example, that cannot be solved for the first paper of almost disturbance decoupling problem, is used to show that the effectiveness of proposed algorithm is better than the singular perturbation method.