Education
- Ph.D. Northwestern University, 2014
- M.S. Northwestern University, 2011
- B.S. Purdue University, 2009
Background
Dr. Aaron Young will be joining Georgia Tech as an Assistant Professor in Mechanical Engineering and is interested in designing and improving powered orthotic and prosthesis control systems for persons with stroke, neurological injury or amputation. He is currently a post-doctoral fellow at the University of Michigan in the Human Neuromechanics Lab working with exoskeletons and powered orthoses to augment human performance. He has previously worked on the control of upper and lower limb prostheses at the Center for Bionic Medicine (CBM) at the Rehabilitation Institute of Chicago. His master’s work at CBM focused on the use of pattern recognition systems using myoelectric (EMG) signals to control upper limb prostheses. His dissertation work at CBM focused on sensory fusion of mechanical and EMG signals to enable an intent recognition system for powered lower limb prostheses for use by persons with a transfemoral amputation.
Research Areas and Descriptors
Automation & Mechatronics and Bioengineering: Upper and lower limb prostheses and orthoses, biological signal processing, intent recognition, control systems based on EMG and mechanical sensors
Control of powered robotics systems
Research
Dr. Young’s research is focused on developing control systems to improve prosthetic and orthotic systems. His research is aimed at helping develop clinically translatable research that can be deployed on research and commercial systems in the near future. Some of the interesting research questions are how to successfully extract user intent from human subjects and use these signals to allow for accurate intent identification. Once the user intent is identified, smart control systems are needed to maximally enable individuals to accomplish useful tasks. For upper limb patients, these tasks might be grasping an object or reaching an arm in space. For lower limb patients, this might be standing from a seated position, walking, or climbing a stair.
Caption: One of the first uses of a full intent recognition system using mechanical and EMG signals helped an amputee climb 103 floors of the Willis Tower. Seamless transitions were enabled between walking and stairs on the powered knee and ankle prosthesis developed by Vanderbilt University.
One of Dr. Young’s primary research interests is in intent recognition for lower limb prostheses and orthoses. Intelligent intent recognition systems are capable of allowing users to perform automatic, seamless and natural transitions between different locomotion modes. These modes include every-day tasks such as sitting, standing, walking, and traversing stairs and slopes. By developing and improving intent recognition and control techniques, we hope to improve the usability of the exciting new mechatronic prostheses and orthoses that are increasingly becoming commercially available.
Caption: Direct prosthesis joint control and advanced timing is possible using EMG signals recorded from the residual limb of a transfemoral amputee. This subject has undergone Targeted Muscle Reinnervation surgery which enables advanced control of the ankle joint as well as the knee.
For upper limb prostheses, pattern recognition control based on myoelectric (EMG) input is promising for extending the number of degrees of freedom. Pattern recognition systems, similar to those used for speech recognition can be applied to an array of EMG inputs from the residual limb of an amputee. These patterns are mapped to specific motions of the prosthesis such as hand grasps or moving an elbow or wrist. One interesting research question that we are pursuing is to use pattern recognition techniques along with other machine learning methods to enable simultaneous movements of multiple joints. Humans in every-day life use simultaneous joint movements to allow for smooth, fluid movements. These types of movements are largely absent in current prosthesis control systems, and the aim is to enable and improve simultaneous joint control through pattern recognition and machine learning techniques.
Caption: Surface EMG signals capture underlying muscle movements. These signals are processed and categorized based on a pattern recognition system which maps overall feature patterns to a specific movement of the prosthesis
Students who work with Dr. Young will work with an interdisciplinary group in robotics, mechanical, electrical and biomedical engineering. They will learn how to conduct human subject experiments and work with clinicians in physical therapy and P&O (Georgia Tech’s Prosthetics and Orthotics Program) to do clinically translatable research. Additionally, they will develop expertise in biological signal processing, mechatronic systems, machine learning, robotics and control.