EM-Sense is a software defined radio solution with the potential to make smartwatches smarter by informing them what a user is touching. By measuring electro-magnetic (EM) signals conducted through the body and interpreted in real time, EM-Sense can quantify the world in new and useful ways. From guiding projects that include sensing what a wearer is doing while providing feedback through to fitness and health tracking, the Internet of Things just became exponentially larger.
Developed by scientists at Disney Research and Carnegie Mellon University, this novel approach to sensing the environment could be game changing since the entire universe of legacy objects can now be detected without costly modifications or the need to add additional sensors.
The premise behind EM-Sense technology is delightfully simple. Our bodies conduct electro-magnetic signals and this conduction can be easily measured. Likewise, objects produce distinctive EM signals of their own. By using the body as an antenna and software to filter the signals, EM-Sense has shown it is possible to discriminate reliably among things as similar as drills and dremels, and to identify large structural features such as doors as well as a wide array of common devices.
What makes EM-Sense particularly exciting is the proof of concept prototype the researchers were able to create using an off-the-shelf software defined receiver. The fact that key components are commonly available likely means quick iteration and rapid uptake into commercial smartwatches and other products.
Though the prototype device is very much a proof-of-concept, this technology holds great potential to bring new layers of functionality to the smartwatch. It could mean that your computer unlocks the moment your watch-wielding wrist is plonked onto its trackpad, for example, or your kettle begins to boil the moment you shut off your alarm clock in the morning.
Benefits of this new technology are obvious from watching the video below, and who knows where else it might lead. Last month, for example, researchers at the University of Washington unveiled a device based on the same principle that aims to track your carbon footprint by monitoring your daily interactions with the world around you.
... though it makes one wonder just how much you want a watch to know about everything you do and everything you touch.
The research paper entitled EMSense: Recognizing Handled, Uninstrumented, Electro-Mechanical Objects Using Software-Defined Radio was presented on November 8 at the ACM Symposium on User Interface Software and Technology (ACM UIST) 2015.
Source: Disney Research, Gierad Laput