Interacción persona-objeto mediante computación ubicua

García Navarro, Diego

Interacción persona-objeto mediante computación ubicua

License

Citation

García Navarro, D. Interacción persona-objeto mediante computación ubicua. 2014. handle: http://hdl.handle.net/10230/22899

This citation was generated automatically.

Abstract

This project aims to detect user activity with objects in the environment in real time in an unassisted manner, to improve processes in commercial environments, using the Radio Frequency Identification (RFID) technology. This work presents a classification system based on Machine Learning that uses information from RFID and Arduino parameters. The goal is to obtain more information on the interaction with the objects that are located in a Smart Shelf. This information would allow a user to take a book from a shelf, and books on similar topics would be displayed on a screen together with a led, visually showing the selection of the book. To improve the accuracy in the classification, the RFID parameters from the main antenna detecting each tag are used. We demonstrate that the system classifies the interaction of objects with 82% accuracy. As a future work, we propose to add more RFID parameters to improve the detection of the interaction, as well as evaluation in a real business environment.