Will the smart kitchen of the future be stocked with arrays of distributed sensors or could a single suite of sensors, localized on a credit-card sized housing, plug into an outlet to imbue the kitchen with all the intelligence it needs? According to Carnegie Mellon researchers in the Future Interfaces Group, the latter concept is highly promising.
The Future Interfaces Group has developed a synthetic sensor-based device that can monitor multiple types of phenomena in a room, including sounds, vibration, light, heat, electromagnetic noise, and temperature. This device, featuring nine sensors, can determine whether a faucet’s left or right spigot is running, if the microwave door is open or how many paper towels have been dispensed.
“The idea is you can plug this in and immediately turn a room into a smart environment,” said Gierad Laput, a Ph.D. student in CMU’s Human-Computer Interaction Institute (HCII). “You don’t have to go out and buy expensive smart appliances, which probably can’t talk to each other anyway, or attach sensors to everything you want to monitor, which can be both hard to maintain and ugly. You just plug it into an outlet.”
Machine learning algorithms combine raw data feeds into powerful synthetic sensors that can identify a wide range of events and objects. For example, they can distinguish between a blender, coffee grinder, and mixer based on sounds and vibrations.
The CMU researchers discuss the technology in the following video and have been demonstrating it at recent conferences:
“Smart appliances are expensive and rarely talk to one another,” the researchers note. “We’ve explored an alternate, general purpose sensing approach where a single, highly capable sensor board can indirectly monitor an entire room. We started our research by taking an inventory of sensors used in commercial and academic systems. Our sensor board is plug-and-play, uses wall power and connects to our cloud over WiFi.”
CMU researchers are also expanding the types of data feeds that the sensors work with. For example, the sensors can infer human activity, such as when someone has left for work, and the sensors can be trained to recognize various phenomena, such as the cycling of heating and air conditioning units. In addition, the sensors can be trained to detect many popular devices and brands of kitchen products
Google, through the GIoTTo Expedition Project, has supported the CMU research, as has the David and Lucile Packard Foundation. Google is also actively pursuing its Tensor Processing Units, or TPUs, which are similar in concept to the general purpose sensors from Carnegie Mellon.
The CMU sensing concepts are, of course, joining many other imaginative new ideas for sensors that could impact smart kitchens. For example, NeOse is a new device that connects to smartphones and databases and can recognize more than 50 types of odors. This smell-sensing device could detect when a food item is spoiled in a refrigerator, when food is being overcooked, and more.
Make sure to check out the Smart Kitchen Summit, the only event about the future of food, cooking, and the kitchen. Use the discount code SPOON to get 25% off of tickets. Also, make sure to subscribe to get The Spoon in your inbox.
Thomas Power says
Why not have Google voice learn by listening to new devices