Robust finite-time nonlinear control of exoskeleton robots in the presence of unknown friction force, parametric sectional number, and bounded external disturbance

Today, many people suffer from physical disability due to various reasons, including war, amputation, or congenital. To help these people, human knowledge has designed a class of robots called exoskeleton robots. These robots are worn by a disabled or weak person to help him to do the desired work without the need of others. A challenging issue related to exoskeleton robots is controlling them to perform the desired action while maintaining coordination with other parts of the person’s body. Also, the modeling of exoskeleton robots is always affected by unfavorable factors such as parametric uncertainties, uncertainty caused by unmodeled dynamics, unknown friction forces, and disturbance. Therefore, it is difficult to design a controller for this type of robot. In recent years, many studies have been conducted on the control problem of exoskeleton robots, but in most of them, the influence of uncertain disturbing factors, friction, and disturbance have not been taken into account. In practical applications, it is necessary for the exoskeleton robot to reach the predetermined paths after a limited time. Therefore, exoskeleton robot design should be done with the aim of time-limited stabilization of the closed-loop system, which most research studies have not paid attention to. In this thesis, the comprehensive nonlinear model for describing the kinematics and dynamics of n-DOF exoskeleton robots is first discussed. It is also assumed that all physical constant parameters (including mass, length, and moment of inertia of the links, the distance of each joint from the center of mass, etc.) are unknown, and the dynamic equations of the model are affected by unknown friction and disturbance forces (with an unknown upper limit). For the said robot, the robust-adaptive nonlinear controller is designed in order to meet the goal of tracking the determined paths and guarantee the time-limited stability of the closed-loop system of the robot. It should be noted that in dealing with the problem of unknown parameters of the model and friction, part of the model is converted into a regression form. The combination of the terminal sliding mode control method and finite-time nonlinear adaptation laws (to estimate the unknown physical parameters of the model, the unknown constants of the friction forces, and the upper limit of disturbance) is used to design the controller. Time-limited stability analysis of the closed-loop exoskeleton robot system is performed using Lyapunov’s fundamental theorem and related lemmas. In the end, the nonlinear controller on the 2-DOF exoskeleton robot is subjected to computer simulation to reveal the correctness of its operation and efficiency.