Gait Recognition Based on Graphene Porous Network Structure Pressure Sensors for Rehabilitation Therapy

Abstract

With the development of health-care and rehabilitation therapy, the collection and analysis of gaits occupy an important position in real-time diagnose. To detect gait patterns accurately, an efficient walking monitoring system is crucial. In this paper, we present a wearable in-shoe system for human gait detection. In our system, a shoe with novel Graphene Porous Network Structure Pressure Sensors (GPNSPS) is used to measure the plantar pressure of walking and an ensemble machine learning method is utilized as the classifier to recognize gait patterns including normal patterns, toe in, toe out, lame feet and heel feet. The flexible and intelligent system demonstrates a promising potential to assist the patients in their rehabilitative care.