Notre Dame engineers Patrick Wensing, an assistant professor of engineering, and second-year Ph.D. student Taylor Gambon are working in collaboration with a company to improve exoskeletons — wearable mobile machines meant to support the body.
“It is a three-year project funded by the National Science Foundation in collaboration with a company called Ekso Bionics that builds assistive exoskeletons based in California,” Wensing said. “The overall goal of the project is to make the exoskeleton more intuitive to operate.”
The exoskeleton currently operates with a control pad, Wensing said, so the goal of the research is to eventually eliminate the need for this control pad and transition to an exoskeleton that is fully understanding of the user’s intentions.
“The exoskeletons that we have are very powerful machines,” he said. “They are strong enough to help you move through your home, to walk up the stairs, to support you while you might be washing dishes, but right now the exoskeletons don’t know when you want to transition between these tasks.”
One way patients utilize exoskeletons is during rehabilitation to help with the patient following a specific gait. Wensing said the primary targets of this research are those with incomplete spinal cord injuries.
“The amazing thing that happens is if you practice your movements through an assistive device or a therapist helping you to walk, your brain will figure out how to rewire what’s left of the spinal cord and then you’ll be back in control of your own motion,” he said. “By making the exoskeleton more intuitive, rather than the exoskeleton controlling you during rehab, you’ll be more in control of your rehabilitation and hopefully more empowered as a result.”
This past summer, Gambon went to California and worked with Ekso Bionics to research the effects of intent changes on exoskeletons, such as intentionally changing walking speed. The exoskeletons were not aware of the intent changes in walking speed, so the results of this research can help the team understand how these changes affect the exoskeleton.
Gambon said she gathered data in a motion capture arena, tracking the changes on both fully functioning individuals and those with spinal cord injuries and watching how both the human and the robot interact with each other. She said she is currently in the process of analyzing this data.
“If we can understand the difference between a normal walking and a speed-up walking or a normal walking and slow-down walking or normal and stop, then hopefully we can help the exoskeleton learn to differentiate the two or several,” Gambon said.
The implications of this intent-change research are more wide-reaching than just in rehabilitation. The results Wensing and Gambon are researching can be used in working with NASA, in prosthesis or in assisting the elderly in the home, Gambon said.
“I think there’s a lot of breadth to this kind of research,” she said. “I don’t think it’s just gonna be pigeonholed to rehabilitation. There are a lot of different areas where this type of intent detection is applicable.”
