October 4, 2022
By Ian Sargent
Associate Professor Aaron Young of the George W. Woodruff School of Mechanical Engineering at the Georgia Institute of Technology has been awarded the prestigious National Institutes of Health (NIH) Director’s New Innovator Award. This is the second NIH award for Young, who was previously awarded the NIH National Center for Medical Rehabilitation Research (NCMRR) Young Investigator Award.
The New Innovator Award is funded through the NIH’s High-Risk, High-Reward Research program. An NIH press release states that the program “catalyzes scientific discovery by supporting highly innovative research proposals that, due to their inherent risk, may struggle in the traditional peer-review process despite their transformative potential.”
“It’s really exciting,” Young said, adding that the award is career-changing. “It provides the funds and scientific freedom to take risks for potentially much higher long-term impact than traditional mechanisms.”
The NIH grant will deliver $2.3 million in funding over five years and will support Young’s research into improving control systems in wearable robotic devices.
“The core innovations are around AI (artificial intelligence) technology and focusing on improving the intelligence of the robotic systems in order to provide a better patient experience using wearable robotic technology,” Young explained.
The clinical applications of Young’s research will transform how wearable robotics restore mobility to stroke populations, amputees, and other individuals with walking impairments. Restoring mobility allows for greater patient independence and can play a major role in improving quality of life. While significant advancements have been made in prosthesis and wearable robotics hardware, optimizing a device for an individual user’s specific needs can be a long and taxing process, and lacks the ability to adapt to different day-to-day scenarios.
Young’s proposal seeks to address this challenge through two key innovations: The first is in AI intent recognition, where the device can be taught to determine what environment the user is in—be it a gentle incline or a hard, slick surface—and what the user is trying to do. By employing adaptive deep learning, where a computer can process data to improve its own performance, a control system can learn to recognize different environments and adjust its behavior accordingly. The adjustments needed, however, will vary from patient to patient, and creating an intelligent agent that can learn an individual’s specific mobility needs without external data input is a central component of Young’s work.
“The goal is to be able to put a device on and, over time, have the system learn your specific gait pattern and adjust its ability to recognize what you do in your daily life,” Young says.
The second area of advancement relates to what Young’s proposal calls the ‘control policy’; this defines how the mechanical components of the device should be moving in order to best provide support for any action in a given environment. By developing an AI system that can self-adapt its control policy without a time-consuming optimization process, wearable robotics can become more personalized and provide more successful patient outcomes.
Young is the director of the Intelligent Prosthetic & Exoskeleton Controls (EPIC) Lab at Georgia Tech, having previously completed a post-doctoral fellowship at the University of Michigan in the Human Neuromechanics Lab and a Ph.D. at Northwestern University in the Center for Bionic Medicine.