Modeling and Gait Tracking of Lower Extremity Exoskeleton using Adaptive Sliding Mode Control Techniques

Abstract

Exoskeletons are orthotic devices powered by technology, aimed at assisting individuals with physical weakness or injuries by enhancing their mobility and facilitating rehabilitation exercises. In industrial and military contexts, exoskeletons are specifically engineered to address the challenge of augmenting human power, enabling soldiers to walk, and run with heavy loads strapped to a backpack frame. This research work deals with the co-simulation of a lower extremity exoskeleton using robust control techniques to improve the performance of lower limbs based on a commercial orthosis using ADAMS and the MATLAB. Kinematic and the dynamic modeling of a six degree of freedom lower extremity exoskeleton is developed and co-simulation is used to examine the resulting models. newlineThe objective of this research is to design, develop, implement, analyze and compare the performance of robust controllers in trajectory tracking of the six joints of the exoskeleton by considering the nonlinear system dynamics. In the presence of uncertainties, parameter variations, and external environmental disturbance, the control system should make the exoskeleton to follow the trajectories of the wearer. The motivation is to develop a robust, reliable, and adaptable control system that ensures exoskeletons can effectively assist or augment human movements, even in the presence of challenges such as nonlinear dynamics, uncertainties, parameter variations, and external disturbances. In this thesis, a super twisting sliding mode controller and an adaptive super twisting sliding mode controller are first tried, and it is seen from the results that the controllers are unable to handle large uncertainties introduced into the system dynamics due to environmental conditions, and that the control effort needed to track the trajectories is large. The control input became continuous, but chattering is not sufficiently reduced, which could harm the actuator used. Therefore, it is suggested that Model Reference Adaptive Control (MRAC)