Unilateral Exoskeleton for Post-stroke Gait Rehabilitation

Christopher Caulcrick

The project aim is to develop a proof of concept unilateral exoskeleton for post-stroke gait rehabilitation and assessment, capable of:

• Tracking reference trajectories and collecting rehabilitation assessment criteria from patient healthy side during exemplar leg swing exercises
• Achieving human-exoskeleton balance during healthy leg swing
• Delivering ‘assistance-as-needed’ during affected leg swing exercises
• Comparing affected side performance against healthy side for evaluating clinical outcomes

Novel engineering methods carried out during the course of this project will include:

• A state estimation approach for recording reference trajectories using inertial data during example rehabilitation exercises
• Proposing and testing metrics for stroke rehabilitation assessment – comparing healthy and affected sides using data from fused inertial and muscle activity data
• Creating model predictive controllers for exoskeleton balance during healthy side leg swing and assistance during affected side leg swing, simulating their implementation, then testing performance on hardware

Control objectives include robustness with respect to modelling uncertainties, trajectory following, overall torque minimisation, external disturbance rejection for stance controller, and, for assistance scenario, external disturbance acceptance for dictating exercise difficulty and delivering ‘assistance-as-needed’.

This research aims to answer the question of unilateral exoskeleton feasibility for post-stroke rehabilitation – identified as a promising means of achieving independent rehabilitation outside of clinical environments.