Researchers at the Massachusetts Institute of Technology (MIT) have created an augmented reality headset that allows wearers to see through objects. Called X-AR, the headset uses computer vision and wireless perception to locate specific items hidden from view, such as inside boxes or under piles. The technology utilises radio frequency (RF) signals, which can pass through materials like cardboard, plastic or wood, to find tagged items. The system displays the location of the item in the headset’s augmented reality interface as a transparent sphere, guiding the user towards it.
Creating X-AR
The researchers had to modify an existing headset with a lightweight antenna capable of communicating with RFID-tagged items. The team used holographic visualisation to display information about the location of the item, presenting the data in a simple way that would be useful in a warehouse or smart home environment. X-AR’s trajectory projection feature presents the path to the tagged object in the form of footprints on the floor.
Applications of X-AR
The technology has several potential applications in retail, warehousing, manufacturing. By locating hidden items, X-AR could reduce time spent searching for inventory or goods, increasing efficiency in warehousing and retail. The technology could also improve safety by detecting objects that are hazardous or hard to find. It could even assist in smart homes.
Looking to the future
The MIT researchers are now exploring how to enhance the technology’s capabilities with different sensing modalities, such as WiFi, mmWave technology or terahertz waves. They are also considering how to extend the headset’s range beyond three meters and enable use by multiple, coordinated headsets. The X-AR headset has taken a significant step forward in the future of AR systems by making them work in non-line-of-sight scenarios, says Ranveer Chandra, managing director of industry research at Microsoft. The research was supported by the National Science Foundation, the Sloan Foundation and the MIT Media Lab.
