tech armband puts your fingers in control
A novel way of interacting with computers and other gadgets could make twiddling your thumbs a productive activity.
Microsoft researchers are developing an armband worn on the forearm that recognises finger movements by monitoring muscle acti asics gel vity. They have called it MUCI, which stands for muscle computer interface.
The aim is to make controlling computers an asics gel d gadgets easier in situations where the user is otherwise engaged for example, when driving a car or taking part in a meeting. Hidden up the user’s sleeve, it lets them control a device with the lift of a finger.
Carefully placed skin electrodes are already used to control prosthetic arms or hands (with video). But accurately placing electrodes in this way is beyond the everyday user without expert help.
Users of MUCI instead just slip the armband onto their forearm, without worrying about how its 10 sensors are positioned. Instead, a series of calibration exercises teach the device’s software how to recognise different movements.
Tests on 10 volunteers showed that, after calibration, the system can recognise the position and pressure o asics gel f all 10 digits with 95% accuracy. It can even detect three different degrees of pressure exerted by a finger.”It works very well if we constrain major arm movements, say, by resting the hand on a desk,” says Microsoft researcher Desney Tan. But MUCI is still “very early stage research”, he adds. More work is needed to make it usable while moving the arm freely.
Giving simple commands using the armband should be easy for example, twitching your pinkie to answer the phone while driving. But the researchers are not yet sure how a user could enter text.
“Morse code is one way, though we hope that is not the eventual solution because it is not easy to learn and it is not amazingly fast,” says Tan, who adds that the efficient typing style used by stenographers could be used.
In the future, the armband could be disguised as a watch or a discrete item of jewellery, say the researchers. “We would like the sensors to be unobtrusive and usable in the real world,” says Tan.
Keep on shopping
Stephen Brewster, an expert in computer interface technology at the University of Glasgow in the UK, is impressed with MUCI.
“Getting 95% accuracy is pretty good for simple gesture recognition,” he told New Scientist. “That would let you get a richer fo asics gel rm of input when you’re out, say, holding heavy shopping bags you could still make some input to a device.”
But MUCI needs to improve if it is to be used for text input, says Brewster.
His own research group is also developing ways to control gadgets without interrupting other activities. Using movement sensors they hope to make gestures like foot tapping, gait changes, or shrugs of the shoulders change tracks on MP3 players or perform other tasks.