Researchers at Hanyang University in South Korea have integrated an electric sensor with a machine learning program, creating a device that can differentiate between surface textures, with potential applications in virtual reality, robotics, and medical prosthetics.
The sensor is fabricated from a graphene-flake film deposited onto a polyethylene naphthalate substrate.
The device registers changes in electrical conductance and resistance via the film when strain causes deformation, boosting the physical contact between individual flakes in the film. The machine-learning program applies the sensor's conductance and resistance data to define specific features connected with different surface texture types.
The researchers have applied the graphene film to an artificial fingerprint structure so it reacts to tiny vibrations caused by the ridges on the fingerprint rubbing against a textured surface; the sensor analyzes these signals to identify the "feel" of differently textured fabrics.
In a blind test of 50 people, the sensor outperformed humans in classifying 12 new fabrics.
From Chemistry World
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found