MIT
Latest
MIT tests autonomous 'Roboat' that can carry two passengers
MIT looked at the original Roboat as “quarter-scale” option, with the Roboat II being half-scale; they’re slowly working up to the point of a full-scale option that can carry four to six passengers.
MIT sleep monitor can track people's sleeping positions using radio signals
A team of MIT researchers has developed a device that can monitor people’s sleep postures without having to use cameras or to stick sensors on their body. It’s a wall-mounted monitor the team dubbed BodyCompass, and it works by analyzing radio signals as they bounce off objects in a room. As the researchers explained, a device that can monitor sleep postures has many potential uses.
MIT researchers use radio signals to detect everyday household activities
The RF-Diary can track someone's activities while maintaining their privacy.
MIT researchers show how 'Dr. Spot' could help diagnose COVID-19
Boston Dynamics and MIT researchers say they’ve collaborated to create “Dr. Dr. Spot also has a computer tablet that allows doctors to remotely speak with patients. Researchers tested Dr. Spot with volunteers at the Harvard Medical School’s Brigham and Women’s Hospital in Boston.
MIT AI system knows when to make a medical diagnosis or defer to an expert
MIT's machine learning system knows when to make a medical diagnosis and when to ask for human assistance.
MIT algorithm finds subtle connections between art pieces
A new system developed by MIT researchers called “MosAIc” is finding hard-to-spot similarities between art pieces at the Metropolitan Museum of Art and Amsterdam’s Rijksmuseum. To use MosAIc, the user inputs an image and MosAIc’s algorithm finds similar art pieces.
MIT researchers create robotic gripper that can untangle thin cables
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory have developed a robotic gripper with the dexterity to handle thin objects like ropes and cables, the university announced.
Autonomous robot uses UVC light to disinfect warehouses
Researchers from MIT have developed a new way to keep shared spaces free of the coronavirus and other pathogens: a UVC light-equipped robot.
AI recreates the painting techniques of famous artists
New AI can recreate the brush strokes used to create a famous painting by looking at the finished image.
MIT fit tens of thousands of artificial brain synapses on a single chip
MIT engineers put tens of thousands of artificial brain synapses on a single chip that's smaller than a piece of confetti.
Study finds security holes in online voting for New Jersey and West Virginia (updated)
Researchers say they've found privacy and security issues in OmniBallot's online voting platform.
MIT gives soft robots a better sense of touch and spatial awareness
Researchers from MIT give robotic grippers a better sense what they’re gripping and how much force to use.
MIT wants to use your microwaving habits to study your health
Researchers from MIT's CSAIL created a wireless system that monitors how people use the appliances in their homes.
Muscle sensors may let you control a drone by clenching your fist
MIT has developed a system that uses muscle sensors to control robots more intuitively.
MIT uses wireless signals and AI to monitor COVID-19 patients at home
A team from MIT's CSAIL developed a wireless device that uses AI to monitor COVID-19 patients in their homes.
MIT is working on a way to track COVID-19 while protecting privacy
MIT's technique could quickly tell if you've been close to a COVID-19 patient without exposing their identities.
MIT project turns spray paint into a functional user interface
Scientists from MIT have developed a way to create interactive surfaces using airbrushed inks -- or in other words, graffiti that actually does stuff. The SprayableTech system lets users create room-sized interactive graphics with sensors and displays that can be applied to everything from walls to furniture. Spray-painted art on walls can be used to turn lights on and off, for example, while designs airbrushed onto the arm of a couch can be used to control a TV. After designing the interactive artwork with a 3D editor, the system generates stencils for airbrushing the layout onto a surface. Then a series of inks are applied -- conductive copper ink, paint, dielectric, phosphor, copper bus and a clear conductor -- and then a microcontroller is attached, connecting the interface to a board that runs the code for sensing and output. The system hinges largely on the stencil design phase, to ensure the inks are placed in the right places and can connect to the microcontroller properly, so at this stage there's not much opportunity for spontaneous exploration -- you couldn't just spray a design on the wall and expect it to control your lights, for example. But the team is now working on creating modular stencils, potentially allowing users to try the system at home without needing to use a 3D editor or cut out stencils themselves. And looking ahead, the team has even bigger plans. Michael Wessely, lead author on the paper about SprayableTech, says, "We view this as a tool that will allow humans to interact with and use their environment in newfound ways. In the future, we aim to collaborate with graffiti artists and architects to explore the future potential for large-scale user interfaces in enabling the internet of things for smart cities and interactive homes."
Robot learns to set the dinner table by watching humans
To date, teaching a robot to perform a task has usually involved either direct coding, trial-and-error tests or handholding the machine. Soon, though, you might just have to perform that task like you would any other day. MIT scientists have developed a system, Planning with Uncertain Specifications (PUnS), that helps bots learn complicated tasks when they'd otherwise stumble, such as setting the dinner table. Instead of the usual method where the robot receives rewards for performing the right actions, PUnS has the bot hold "beliefs" over a variety of specifications and use a language (linear temporal logic) that lets it reason about what it has to do right now and in the future.
AI can help find more places to store captured CO2 underground
Scientists dream of storing captured CO2 emissions underground, but it can be tricky to find pockets for storing those emissions -- Earth's seismic "hum" makes it difficult to spot the low-frequency waves that would best help map the subterranean landscape. That might not be a challenge in the future thanks to AI. MIT scientists have developed a machine learning system that uses earthquakes to accurately estimate those low-frequency waves and map underground structures. The trick, according to researchers, was in the training.
MIT helps self-driving cars ‘see’ through snow and fog
Self-driving technology has come a long way, but it can still be tripped up by bad weather. A team from MIT's Computer Science and Artificial Intelligence Lab (CSAIL) may have a solution. They've developed a way to help autonomous vehicles "see" by mapping what's beneath the road using Ground Penetrating Radar (GPR).