Theses Doctoral

Building Better Exoskeletons: Understanding How Design Affects Robot Assisted Gait Training

Stegall, Paul

Physical therapy is a field with ever increasing demands as the population ages, resulting in a larger number of individuals living with impairments. Therapy is both physically intensive and time intensive for physical therapists, and can require more than one therapist per patient. The use of technology can reduce both these physical and time demands if appropriately applied, while improving repeatability and providing quantitative evaluation of performance. Through these abilities, it may also improve the quality of life for patients. The work presented here explores how the mechanical and controller design of exoskeletons can be used to improve adaptations to new gait patterns in healthy individuals. Armed with this knowledge, new treatment methods can be adapted, applied, and validated for impaired populations with the intention of recovering a more natural gait pattern.
First, the ALEX II device is presented. It is a unilateral device, designed to aid in gait training for stroke survivors. The previous version, ALEX I, had several limitations in terms of pelvic freedom, leg range of motion, and the support of the gravitational load. ALEX II was designed to address these issues. Next, a study is presented, using healthy young adults (N=30), in which ALEX II was used to explore how the amount of freedom allowed at the pelvis during gait training affects the level of adaptation subjects are able to achieve. This was evaluated for five separate configurations which resemble existing exoskeletons. It was found that intermediate levels of pelvic freedom degrade the amount of adaptation and that pelvic translation contributes more to this effect than hip abduction/adduction.
The next work concerns the design of ALEX III, a bilateral device with twelve active degrees-of-freedom. ALEX III was created to increase the ability to explore the functionality required for gait training, which is why it is capable of controlling 4 degrees-of-freedom at each leg, and 4 degrees-of-freedom at the pelvis. This is followed by the the design of a new type of haptic feedback which utilizes a variable, viscous damping field, which increases the damping coeffiecent as the subject moves away from a specified path. This feedback type was tested in a set of experiments in healthy young adults. The first study (N=32) compared four different settings for the new feedback, finding that while all groups demonstrated adaptations in gait, the lowest rate of change of the damping field exhibited less adaptation. The final study (N=36) compared this haptic feedback to two previously used haptic feedback types. The previously used feedback strategies used a force that pushed the leg either towards or away from the desired path. All three of these strategies were found to produce similar levels of adaptation, however the damping field used much less external force. These findings may change the way exoskeletons for gait training are designed and increase their accessibility.
While all the findings need to be validated in impaired populations they can still inform the design of future exoskeletons. The first finding, that providing an intermediate amount of freedom to the pelvis can interfere with gait training, suggests that future devices should have very high amounts of freedom or very restricted pelvic motions. The final finding, that damping fields can be used to induce gait adaptations using a much lower force, can drastically change exoskeleton design and how robotic therapy is provided. Exoskeletons can be made lighter as a result of the force being highly reduced so that lighter weight components can be used, and the dissipative nature of the force reduces dependence on heavy power sources because regenerative breaking can be used to power the device. These factors also make it possible to for devices to be used overground, which may make training more transferable to the real world.


  • thumnail for Stegall_columbia_0054D_13596.pdf Stegall_columbia_0054D_13596.pdf application/pdf 18.4 MB Download File

More About This Work

Academic Units
Mechanical Engineering
Thesis Advisors
Agrawal, Sunil K.
Ph.D., Columbia University
Published Here
October 6, 2016