University of Pittsburgh researchers receive award to develop ultrasonic sensors for an exoskeleton

The National Science Foundation’s Cyber-Physical Systems program has awarded researchers from the University of Pittsburgh a 3-year $400,000 grant to develop an ultrasound sensor system for a hybrid exoskeleton, which would use both electrical nerve stimulation and external motors.

According to a press release, the University of Pittsburgh researchers will collaborate with colleagues at George Mason University, who have also received a $400,000 award from the National Science Foundation.

“One of the most serious impediments to developing a human exoskeleton is determining how a person who has lost gait function knows whether his or her muscles are fatigued,” Nitin Sharma, PhD, MS, assistant professor of mechanical engineering and materials science at the University of Pittsburgh Swanson School of Engineering, said in the release. “An exoskeleton has no interface with a human neuromuscular system, and the patient does not necessarily know if the leg muscles are tired, and that can lead to injury.”

The research group will also develop computational algorithms for real-time sensing of muscle function and fatigue. Human subjects using a leg-extension machine will allow for detailed measurement of strain rates, transition to fatigue and full fatigue to create a novel muscle-fatigue prediction model. Future phases will allow the teams to develop a wearable device for patients with motor impairment.

“Right now an exoskeleton combined with ultrasound sensors is just a big machine, and you do not want to weigh down a patient with a backpack of computer systems and batteries,” Sharma said in the release. “The translational research with George Mason will enable us to integrate a wearable ultrasound sensor with a hybrid exoskeleton, and develop a fully functional system that will aid in rehabilitation and mobility for individuals who have suffered spinal cord injuries or strokes.”


Leave a Reply

Your email address will not be published.