This Robot Librarian Locates Haphazardly Placed Books
It’s the next best thing since the Dewey Decimal System
/https://tf-cmsv2-smithsonianmag-media.s3.amazonaws.com/filer/17/fb/17fb67ef-aa5a-4cf6-b2f8-af5f24bbefcd/screen_shot_2016-06-10_at_13146_pm.png)
Organization rules in the library stacks, but patrons can easily thwart the system by haphazardly returning books to the shelves. Librarians spend many hours searching for these wandering tomes, but robots could soon help them out. A new librarian robot locates misplaced books, helping to return them to their rightful place, Coby McDonald writes for Popular Science.
Over the years, automation has slowly crept into libraries around the world. Digital databases replaced card catalogs, and some libraries use robots to file, sort and retrieve books for patrons. But most local libraries lack the space and resources for such complex systems.
Enter AuRoSS, the robot librarian.
A group of researchers at Singapore’s Agency for Science, Technology and Research (A*STAR) have developed a robot that can wander among the stacks at night, scanning shelves for misplaced books. When the Autonomous Robotic Shelf Scanning system (AuRoSS) finds one, it flags it so a librarian can go back later to grab the book and return it to where it belongs.
In order to identify and keep track of the books, AuRoSS relies on Radio Frequency Identification (RFID) tags. These little chips are used in everything from office key cards to passports. In recent years, libraries have begun using them to help keep track of books, scanning the spines with hand-held devices. But AuRoSS can trundle around the stacks all on its own, continually scanning the little tags, according to McDonald.
The stacks, however, can become complex labyrinth, challenging for human navigation, let alone robotic. In order for AuRoSS to successfully scan RFID tags, it has to stay at just the right distance away from the shelves. “Too far and we lose the RFID signals, but too close and the antenna hits the shelf,” project leader Renjun Li says in a statement.
At the same time, library maps are often too low-resolution to be useful for robots. While basic maps direct patrons to sci-fi/fantasy books from the history section, robots require extremely precise detail and directions for everything they do.
So Li’s team programmed AuRoSS to detect the surface of the bookshelves when planning out its route. By attaching the RFID-detecting antenna and a set of ultrasonic scanners to a robotic arm, AuRoSS can keep its sensors close enough to detect the books and know when it needs to change direction to continue scanning shelves.
During a recent demonstration at Singapore’s Pasir Ris Public Library, Li’s team found that AuRoSS could navigate the library and detect misfiled books with 99 percent accuracy. While AuRoSS still requires some refining, it has the potential to take on some of a librarian's most tedious tasks.