A wearable device engineered by researchers from University of California Berkeley, IBM Research Zurich and University of Bologna (Italy) recognizes hand gestures based on electrical signals detected in the forearm. The system, which combines biosensors with artificial intelligence (AI), could one day be used to control prosthetics or to interact with almost any type of electronic device.
The device that combines wearable biosensors with AI software to help recognize what hand gesture a person intends to make based on electrical signal patterns in the forearm. Source: University of California BerkeleyThe flexible device recognizes hand gestures based on 64 different electrical signals in the forearm. The signals are fed into a chip programmed with hyperdimensional computing algorithm trained to distinguish 21 individual hand gestures, including a thumbs-up, a fist, a flat hand, holding up individual fingers and counting numbers.
The AI software algorithm “learns” how electrical signals in the arm correspond with individual hand gestures — done by wearing the cuff and making hand gestures one by one. When a user eventually makes one of these gestures or even thinks about doing so, the AI system is able to determine which one it is by matching its distinct nerve signal pattern up with one already learned. The algorithm also automatically updates to compensate for new variables such as an arm being held in an unusual position. All of the processing takes place within the chip, so no user data is transmitted to the cloud.
The gesture recognition and interpretation device is expected to pave the way for better prosthetic control and seamless interaction with electronic devices.
The research is published in Nature Electronics.
