A Combined Finite-Time Control Framework for Exoskeleton Robots by Utilizing Adaptive-Robust Nonlinear Control Method

In this study, by using the adaptive-robust nonlinear control approach, an innovative hybrid finite-time control framework is introduced to tackle the tracking problem for a great group of exoskeleton robots (in the presence of friction forces, two types of uncertainties, and unknown forces generated by the disabled person). According to the tracking aim, angular displacement of robot must exactly tend to required trajectories within the finite time. Firstly, a general nonlinear model is represented to characterize dynamical behavior of a typical exoskeleton robot possessing unknown physical constants. To complete this model, friction forces, modelling uncertainties, and unknown human torques (external disturbances) are considered. Two components of the exoskeleton model (unknown friction forces and parametric uncertainties) are rewritten as two detached linear regression forms. Secondly, a finite-time adaptive-robust nonlinear control structure is proposed to accomplish the aforementioned tracking aim and, as a result, the global finite-time stability is provided for the closed-loop exoskeleton robot. The mentioned finite-time nonlinear controllers are designed by combining the adaptation rules and the terminal sliding mode control strategy (along with new defined sliding manifolds). These adaptation rules estimates model physical constants, unknown coefficients of the friction forces, unknown human torques and upper bound of the Euclidean norm of the external disturbance vector. Mathematical analysis illustrates that time responses of the estimations precisely tend to constant values after the finite-time. Eventually, the combined finite-time control framework is simulated onto the 2-DOF exoskeleton numerically and obtained results reveal that the proposed control structure appropriately provides the finite-time tracking objective. the maximum power point tracking objective.