May 28, 2013

UIUC Student Developing an Algorithm to Drive Prosthetic Controllers

Content provided by The O&P EDGE
Current Issue - Free Subscription - Free eNewsletter - Advertise

Navid Aghasadeghi, MS, a doctoral student in electrical and computer engineering at the University of Illinois at Urbana-Champaign (UIUC), is developing an algorithm that will help provide control to people who use computerized lower-limb prostheses and allow them to walk with a natural gait.

The prosthetic device being used to test Aghasadeghi’s algorithm was developed at Vanderbilt University, Nashville, Tennessee. Photograph courtesy of UIUC Coordinated Science Lab.

“With powered devices, if we can provide the right control to the device, we should be able to provide natural locomotion to the amputee,” said Aghasadeghi. Aghasadeghi’s research began by observing how unimpaired humans walk and then he attempted to replicate that using a prosthetic device. He is using a theory called inverse optimal control, which hypothesizes that human motor control (moving an arm or leg, etc.) can be modeled as being optimal with respect to a performance criterion. He created an algorithm that uses those optimal criteria and the height and weight of the prosthetic user to derive controllers for prosthetic devices.

“Every amputee has to go into a clinic where the clinician chooses the parameters,” Aghasadeghi said. “The [amputee] would walk with the device, try out the gait, and [the clinician would] tweak the parameters. We’re trying to customize the controller parameters for the amputees, so clinicians don’t have to spend hours figuring out what the parameters should be. They can just use our algorithm….”

Testing of the algorithm has been done on two able-bodied subjects walking on flat ground. Aghasadeghi has plans to test the algorithm on prosthetic users as well as in different locomotion modes, such as walking up stairs or down ramps.

Editor’s note: This story was adapted from materials provided by the UIUC Coordinated Science Lab.

Bookmark and Share