We are searching data for your request:
Upon completion, a link will appear to access the found materials.
MIT has developed a system that uses RFID tags to help robots find objects with more speed and accuracy. The novel idea could be applied to robots working in packaging and assembly as well as search and rescue drones.
SEE ALSO: FIFTY EMPLOYEES WILL GET MICROCHIP IMPLANTS AT THIS COMPANY'S INAUGURAL 'CHIP PARTY'
In a presentation, the researchers demonstrated that the robots can locate tagged objects within 7.5 milliseconds, with an error factor of less than a centimeter.
The system, dubbed TurboTrack, works by first placing an RFID (radio-frequency identification) tag on an object. The reader sends a wireless signal that bounces off the RFID tag and other nearby objects and rebounds to the reader.
An algorithm sorts through the signals to find the RFID’s response. TurboTrack's inventors say the system could replace computer vision in some instances. Just like humans, computer vision is limited by how well it can see, and it can miss objects in cluttered environments.
Radio signals could eliminate the need for computer vision
Radio signals face no such challenges they can find objects no matter the clutter or visual impairment. To test the system, the researcher placed an RFID tag onto a bottle cap and another to its respective bottle.
A robotic arm was able to locate the cap and place it on the bottle, which was being held by another robotic arm. In another validation, the researchers tracked RFID-equipped nano drones during docking, maneuvering, and flying.
In both examples, Turbo Track was as fast and as accurate as traditional computer-vision systems.
“If you use RF signals for tasks typically done using computer vision, not only do you enable robots to do human things, but you can also enable them to do superhuman things,” says Fadel Adib, an assistant professor and principal investigator in the MIT Media Lab, and founding director of the Signal Kinetics Research Group.
“And you can do it in a scalable way because these RFID tags are only 3 cents each.”
Cheap vision good answer for assembly lines
The system could be easily applied to manufacturing where robots are used in assembly. Robots that need to pick up, assemble and package items along an assembly line could use the RFID system instead of more expensive computer vision.
Another excellent application for the TurboTrack system would be on handheld “nano-drones” used in search and rescue mission. The nano drones currently use computer vision to stitch together captured images for localization purposes.
But the tiny drones can easily lose each other and get confused by architecture. This limits their ability to coordinate and spread out over a larger area and collaborate to search for a missing person or object.
However, using TurboTrack the swarms could better locate each other.
“You could enable a swarm of nanodrones to form in certain ways, fly into cluttered environments, and even environments hidden from sight, with great precision,” says first author Zhihong Luo, a graduate student in the Signal Kinetics Research Group.
The researchers will present their paper next week at the USENIX Symposium on Networked Systems Design and Implementation.