An RFID framework for multimodal service provision

In recent years there has been a growing interest toward the development of pervasive and context-aware services, and RFID technology played a relevant role in the context sensing task. We propose the use of RFID technology together with a conversational agent in order to implement a multimodal information retrieval service we call SensorMesh. The information acquired from RFID tags about the nearest point of interest is processed by the conversational agent that carries a more natural interaction with the user, also exploiting a common sense ontology. The service is accessible using a multimodal browser on Personal Digital Assistants (PDAs); the browser allows the user to interact with the conversational agent by means of spoken language instead of the traditional, keyboard- (or stylus-) based input systems. The resulting system offers a more natural interaction with respect to traditional prerecorded, audio-visual set-vices, and it is particularly suitable for non technology-skilled users.