Intelligent Control and Sensing Interfaces for a Complaint Knee Exoskeleton Toward Enhanced Human-Robot Interaction Dynamics

Abstract

Research in compliant control and actuation for lower limb exoskeletons is crucial. This thesis emphasizes the critical domain of compliant control and actuation for lower limb exoskeletons, emphasizing the development of adaptive gait support mechanisms to enhance user mobility and autonomy. Despite the increase of control techniques within the extant literature, a conspicuous research void persists concerning the integration of compliant actuators in exoskeletons, especially for achieving personalized gait assistance essential for optimizing mobility support. Furthermore, the absence of established benchmark criteria for comparing the efficacy and performance of these systems underscores the need for a standardized evaluation framework. Addressing these gaps, this work introduces multiple control strategies for a knee exoskeleton, coupled with the integration of an innovative compliant actuator, and proposes a comprehensive benchmarking schema to evaluate the system's performance through both quantitative and qualitative metrics. Research methodologies encompass simulations via MATLAB/Simulink and empirical experiments conducted in the ARIC lab at Khalifa University. This study culminates in an analytical discourse on the assessment outcomes, offering insights for refinement and delineating the limitations of the proposed approaches. The development and implementation of multiple control strategies for a knee exoskeleton, including trajectory tracking, torque, and impedance controls, have been successfully validated through rigorous simulations and empirical experiments. These methodologies have showcased the exoskeleton's ability to adapt to user-specific gait patterns and provide personalized assistance, which is critical for enhancing the autonomy and mobility of individuals with lower limb disabilities.

Date of Award	30 Apr 2024
Original language	American English
Supervisor	Irfan Hussain (Supervisor)