This Keyboard Can Be Crumpled Like a Handkerchief
Matthew Greenwood posted on July 02, 2018 |

While portable keyboards are already on the market, they aren’t very flexible and are still bulky even when folded up. But a group of South Korean researchers may have just created a truly portable keyboard—one you can scrunch up and stick in your pocket.

The team of engineers, from the Faculty of Nanotechnology and Advanced Materials Engineering at Sejong University in Seoul, made a keyboard out of a sheet of soft silicone rubber. The material is embedded with conductive carbon nanotubes that respond to the push of a finger by changing electrical resistance. The designers then drew squares that recreated a keyboard’s letters, numbers and characters on the sheet.

The researchers used a machine learning technique based on deep neural networks to train the keyboard to identify the letter or character being typed. This was done by recording the changes in electrical resistance that occur when the mat is pressed, and the precise location of those presses. That data was recorded at several locations on the edge of the sheet, along with a huge number of random touches. This allowed the keyboard to recognize patterns in the data, giving it the ability to differentiate actual keystrokes from random contact.

The team stated that their simple keyboard worked perfectly and outperformed all existing portable keypads in terms of functionality, flexibility, disposability and cost.

Existing portable keyboards use rigid push buttons, foldable sheets or textured pads; these options use brittle components and are expensive to produce. Using a simple one-piece sheet of silicone rubber would make it much cheaper and quicker to produce. In fact, the keyboard is expected to cost only a dollar, essentially making it disposable.

The aim of the team was to create a truly portable keyboard that could stand up to the rigors of everyday life. Their creation promises to be able to take a beating, and be cheap enough to replace when it wears out.

It also has the potential to allow for computing in difficult environments where size and weight restrictions, and durability, are important factors—such as for explorers in remote locations or emergency responders in a natural disaster.

Read more about how machine learning is improving consumer electronics at No More “Nuisance Trips” with Smart Power Outlets for Smart Homes.


Recommended For You