According to the vision of Ambient Intelligence (AmI), the most advanced technologies are those that disappear: at maturity, computer technology should become invisible. All the objects surrounding us must possess sufficient computing capacity to interact with users, the surroundings and each other. The entire physical environment in which users are immersed should thus be a hidden computer system equipped with the appropriate software in order to exhibit intelligent behavior. The objective of our research is to take steps in this direction by proposing a feasible software application able to learn the behavior and habits of home inhabitants in order to anticipate their needs. The result is an adaptive, context-aware application that works as an integral part of a dedicated middleware designed to make today's heterogeneous, mostly incompatible domotic systems fully interoperable. By applying machine learning techniques, it offers a complete, ready-to-use practical application that learns through interaction with the user in order to improve quality of life in a technological living environment, such as a house, a smart city and so on. Although the proposed solution is currently suitable for application to comfort issues, it also represents an opportunity to provide greater autonomy and safety to disabled and elderly occupants, especially the critically ill, as its results can be utilized in applications that can recognize unusual, potentially hazardous situations. In fact, this modeling process serves as the basis for implementation of an e-health system that can actively contribute to anticipating, and thereby preventing, emergency situations. The prototype has been developed and is currently running at the Pisa CNR laboratory, where a home environment has been faithfully recreated.

An AmI approach to anticipate home inhabitant's needs

Miori V;Russo D
2016

Abstract

According to the vision of Ambient Intelligence (AmI), the most advanced technologies are those that disappear: at maturity, computer technology should become invisible. All the objects surrounding us must possess sufficient computing capacity to interact with users, the surroundings and each other. The entire physical environment in which users are immersed should thus be a hidden computer system equipped with the appropriate software in order to exhibit intelligent behavior. The objective of our research is to take steps in this direction by proposing a feasible software application able to learn the behavior and habits of home inhabitants in order to anticipate their needs. The result is an adaptive, context-aware application that works as an integral part of a dedicated middleware designed to make today's heterogeneous, mostly incompatible domotic systems fully interoperable. By applying machine learning techniques, it offers a complete, ready-to-use practical application that learns through interaction with the user in order to improve quality of life in a technological living environment, such as a house, a smart city and so on. Although the proposed solution is currently suitable for application to comfort issues, it also represents an opportunity to provide greater autonomy and safety to disabled and elderly occupants, especially the critically ill, as its results can be utilized in applications that can recognize unusual, potentially hazardous situations. In fact, this modeling process serves as the basis for implementation of an e-health system that can actively contribute to anticipating, and thereby preventing, emergency situations. The prototype has been developed and is currently running at the Pisa CNR laboratory, where a home environment has been faithfully recreated.
2016
Istituto di Scienza e Tecnologie dell'Informazione "Alessandro Faedo" - ISTI
Ambient intelligence
Domotics
Home automation
DomoNet
SOA
AAL
Ontology
Artificial intelligence applications and expert systems
File in questo prodotto:
File Dimensione Formato  
prod_359026-doc_117652.pdf

accesso aperto

Descrizione: An AmI approach to anticipate home inhabitant's needs
Dimensione 1.27 MB
Formato Adobe PDF
1.27 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14243/318582
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact