The home automation systems of tomorrow won’t look anything like they do right now. In the future, you wont have to whip out your smartphone and fire up an app to flip your lights on. Hell, you probably wont even need a smartphone.
Instead, the automated home of the future will look much more like JARVIS from Iron Man. No clunky remote controls or proximity-based Bluetooth tags – just wave your arms, point at whatever you want to interact with, an always-on 3D motion detector will sense your movements and make it happen.
These types of systems are already in the works. Companies like PointGrab and Frog Designs have built working prototypes that allow you to control your home appliances via gestures, but these systems rely on cameras to detect your motion – a technique that’s proven difficult, unwieldy, and unreliable.
A team of engineers from MIT has devised a better method – one that eschews camera-based detection in favor of radio waves, and can detect motion in three-dimensional space anywhere in the house, even through walls.
The technology, dubbed WiTrack, uses radio signals “whose power is 100 times smaller than Wi-Fi and 1,000 times smaller than cellphone transmissions.” These radio signals reflect off of moving human bodies, and with the help of receivers and triangulation algorithms, the system can use them to effectively determine a person’s position inside the house with an accuracy between 10 and 21 centimeters.
Furthermore, in addition to tracking your position, WiTrack can also detect the motion of individual limbs. This means it can not only be used for things like turning lights on/off when you enter or leave a room, but also for specific gesture controls like pointing at appliances to switch them on. Imagine a future where your house can tell when you’ve gotten out of bed in the morning and automatically turn on the hot water in the shower, or send you an emergency alert when grandma takes a spill in the living room. A system like WiTrack would make stuff like that possible.
Right now the system can only track one person at a time, but it serves as solid proof-of-concept that JARVIS-style whole-home gesture control is possible without retrofitting every room with an array of cameras and sensors. Right now it’s just an academic research project, but it’s highly likely to blossom into something bigger in the next few years.
