Ambient Intelligence is new multidisciplinary paradigm that is going to change the relation between humans, technology and the environment they live in. This paradigm has its roots in the ideas Ubiquitous and Pervasive computing. In this vision, that nowadays is almost reality, technology becomes pervasive in everyday lives but, despite its increasing importance, it (should) becomes “invisible”, so deeply intertwined in our day-to-day activities to disappear into the fabric of our lives. The new environment should become “intelligent” and “smart”, able to actively and adaptively react to the presence, actions and needs of humans (not only users but complex human being), in order to support daily activities and improve the quality of life. Ambient Intelligence represents a trend able to profoundly affect every aspect of our life. It is not a problem regarding only technology but is about a new way to be “human”, to inhabit our environment, and to dialogue with technology. But what makes an environment smart and intelligent is the way it understands and reacts to changing conditions. As a well-designed tool can help us carry out our activities more quickly and easily, a poorly designed one could be an obstacle. Ambient Intelligence paradigm tends to change some human’s activities by automating certain task. However is not always simple to decide what automate and when and how much the user needs to have control. In this thesis we analyse the different levels composing the Ambient Intelligence paradigm, from its theoretical roots, through technology until the issues related the Human Factors and the Human Computer Interaction, to better understand how this paradigm is able to change the performance and the behaviour of the user. After a general analysis, we decided to focus on the problem of smart surveillance analysing how is possible to automate certain tasks through a context capture system, based on the fusion of different sources and inspired to the paradigm of Ambient Intelligence. Particularly we decide to investigate, from a Human Factors point of view, how different levels of automation (LOAs) may result in a change of user’s behaviour and performances. Moreover this investigation was aimed to find the criteria that may help to design a smart surveillance system. After the design of a general framework for fusion of different sensor in a real time locating system, an hybrid people tracking system, based on the combined use of RFID UWB and computer vision techniques was developed and tested to explore the possibilities of a smart context capture system. Taking this system as an example we developed 3 simulators of a smart surveillance system implementing 3 different LOAs: manual, low system assistance, high system assistance. We performed tests (using quali-quantitative measures) to see changes in performances, Situation Awareness and workload in relation to different LOAs. Based on the results obtained, is proposed a new interaction paradigm for control rooms based on the HCI concepts related to Ambient Intelligence paradigm and especially related to Ambient Display’s concept, highlighting its usability advantages in a control room scenario. The assessments made through test showed that if from a technological perspective is possible to achieve very high levels of automation, from a Human Factors point of view this doesn’t necessarily reflect in an improvement of human performances. The latter is rather related to a particular balance that is not fixed but changes according to specific context. Thus every Ambient Intelligence system may be designed in a human centric perspective considering that, sometimes less can be more and vice-versa.

New Challenges in HCI: Ambient Intelligence for Human Performance Improvement

-
2012-03-06

Abstract

Ambient Intelligence is new multidisciplinary paradigm that is going to change the relation between humans, technology and the environment they live in. This paradigm has its roots in the ideas Ubiquitous and Pervasive computing. In this vision, that nowadays is almost reality, technology becomes pervasive in everyday lives but, despite its increasing importance, it (should) becomes “invisible”, so deeply intertwined in our day-to-day activities to disappear into the fabric of our lives. The new environment should become “intelligent” and “smart”, able to actively and adaptively react to the presence, actions and needs of humans (not only users but complex human being), in order to support daily activities and improve the quality of life. Ambient Intelligence represents a trend able to profoundly affect every aspect of our life. It is not a problem regarding only technology but is about a new way to be “human”, to inhabit our environment, and to dialogue with technology. But what makes an environment smart and intelligent is the way it understands and reacts to changing conditions. As a well-designed tool can help us carry out our activities more quickly and easily, a poorly designed one could be an obstacle. Ambient Intelligence paradigm tends to change some human’s activities by automating certain task. However is not always simple to decide what automate and when and how much the user needs to have control. In this thesis we analyse the different levels composing the Ambient Intelligence paradigm, from its theoretical roots, through technology until the issues related the Human Factors and the Human Computer Interaction, to better understand how this paradigm is able to change the performance and the behaviour of the user. After a general analysis, we decided to focus on the problem of smart surveillance analysing how is possible to automate certain tasks through a context capture system, based on the fusion of different sources and inspired to the paradigm of Ambient Intelligence. Particularly we decide to investigate, from a Human Factors point of view, how different levels of automation (LOAs) may result in a change of user’s behaviour and performances. Moreover this investigation was aimed to find the criteria that may help to design a smart surveillance system. After the design of a general framework for fusion of different sensor in a real time locating system, an hybrid people tracking system, based on the combined use of RFID UWB and computer vision techniques was developed and tested to explore the possibilities of a smart context capture system. Taking this system as an example we developed 3 simulators of a smart surveillance system implementing 3 different LOAs: manual, low system assistance, high system assistance. We performed tests (using quali-quantitative measures) to see changes in performances, Situation Awareness and workload in relation to different LOAs. Based on the results obtained, is proposed a new interaction paradigm for control rooms based on the HCI concepts related to Ambient Intelligence paradigm and especially related to Ambient Display’s concept, highlighting its usability advantages in a control room scenario. The assessments made through test showed that if from a technological perspective is possible to achieve very high levels of automation, from a Human Factors point of view this doesn’t necessarily reflect in an improvement of human performances. The latter is rather related to a particular balance that is not fixed but changes according to specific context. Thus every Ambient Intelligence system may be designed in a human centric perspective considering that, sometimes less can be more and vice-versa.
6-mar-2012
HCI
Interazione uomo-macchina
ambient intelligence
fattori umani
human factors
localization technologies
localizzazione
Dibitonto, Massimiliano
File in questo prodotto:
File Dimensione Formato  
PhD_Dibitonto_Massimiliano.pdf

accesso aperto

Tipologia: Tesi di dottorato
Dimensione 3.11 MB
Formato Adobe PDF
3.11 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/11584/266056
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact