The UJI Aerial Librarian Robot: A Quadcopter for Visual Library Inventory and Book Localisation

Abstract

Over time, the field of robotics has provided solutions to automate routine tasks in differentscenarios. In particular, libraries are awakening great interest in automated tasks since they aresemi-structured environments where machines coexist with humans and several repetitive operationscould be automatically performed. In addition, multirotor aerial vehicles have become very popularin many applications over the past decade, however autonomous flight in confined spaces stillpresents a number of challenges and the use of small drones has not been reported as an automatedinventory device within libraries. This paper presents the UJI aerial librarian robot that leveragescomputer vision techniques to autonomously self-localize and navigate in a library for automatedinventory and book localization. A control strategy to navigate along the library bookcases ispresented by using visual markers for self-localization during a visual inspection of bookshelves. Animage-based book recognition technique is described that combines computer vision techniques todetect the tags on the book spines, followed by an optical character recognizer (OCR) to convert thebook code on the tags into text. These data can be used for library inventory. Misplaced books can beautomatically detected, and a particular book can be located within the library. Our quadrotor robotwas tested in a real library with promising results. The problems encountered and limitation of thesystem are discussed, along with its relation to similar applications, such as automated inventoryin warehouses.