The aim of the technology is to improve the use of robots in the retail sector, leading to faster and more accurate packaging and assembly. The technology may also assist drones in terms of cooperative activities for swarming or for engaging in search and rescue missions.
In trials, the U.S. scientists successfully showed that robots equipped with the technology can track down tagged objects within a precision of 7.5 milliseconds. The calculated error rate stands at less than one centimeter.
To test out the technology, the scientists looked at a packing robot and instructed it to move a cap onto a bottle. Based on this, it would be possible, with tagged objects that are hidden, to locate such items on warehouse shelves.
The research group also tracked RFID-equipped nanodrones for activities like docking, maneuvering, and flying, simulating how a drone might be used in a warehouse.
To achieve these feats, the robots and drones have been equipped to read radio-frequency identification (RFID) tags from tens of meters away and by identifying the tags’ locations. This enables widespread coverage across warehouses, for example.
The developed technology is called TurboTrack, and it can be applied to almost any type of object. The process works by a reader sending a wireless signal. The signal reflects off the RFID tag and other nearby objects, and rebounds back to the reader-sensor. An proprietary algorithm then works through all the reflected signals in order to detect the RFID tag response.
Following this, the device undertakes computations which assess the RFID tag’s movement, which serves to boost localization accuracy.
READ MORE: Using drones in large warehouses to prevent inventory mismatches
The researchers, according to Laboratory Manager magazine, see the technology as being superior to computer visions for robots. With computer vision this has the same limitations as human vision, in that it can fail to detect objects in tight or cluttered spaces. The use of wireless signals overcomes this limitation.
According to lead researcher Fadel Adib: “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. And you can do it in a scalable way, because these RFID tags are only 3 cents each.”
The research has been presented at the 2019 USENIX Symposium on Networked Systems Design and Implementation. In the research brief, the scientists explore applications around improving how robots could work on packing lines or for assembling components.
The paper also discusses drone applications, both with drones active in the retail space and with drones that could be used for search and rescue missions. By working in tandem through radio signals, the accuracy of search missions would increase especially where the terrain is chaotic and difficult for a drone equipped with conventional technology to ‘visualise’.
Essential Science
This article is part of Digital Journal’s regular Essential Science columns. Each week Tim Sandle explores a topical and important scientific issue. Last week we looked at a new concern with the antimicrobial triclosan, which is added to many consumer products. This is due to an association with antimicrobial resistance.
The week before we investigated a new technology that can convert plastic waste (the biggest environmental issue associated with modern society) into a clean fuel. This represents one useful application to address the global plastic crisis.
