Considering that our society is aging and the number of elderly people is increasing every day, it is an ethical imperative to develop technologies tailored to their needs. The context of this thesis was to explore the integration of speech-based technologies, socially assistive robots (SARs), and IoT to support the elderly in everyday life at home or care center. Specifically, a speech-based interface has been developed for social robots connected with smart space sensors and actuators, enabling robots to: - engage the elderly in context-aware conversations related to everyday tasks or to behaviors that are consistent with indoor and outdoor environmental conditions - to act as a proxy for domotic control The technological solution has been designed to be as much as possible robot independent and compatible with different robotic systems. We ran two small scale experimental studies (5 participants each), the first aimed at understanding the perceived complexity of using a speech-based system in domestic environments and the second focused on the impact embodiment and movement have on human-computer interaction in said environments. Concerning the first study, results of NASA-TLX questionnaires reveal that the system should be sped up and speech recognition should be improved. Results of the second study highlight the increased engagement and social presence of embodied system able to move, compared to static ones.
La nostra società sta invecchiando e il numero di persone anziane aumenta ogni giorno. É quindi un imperativo etico sviluppare tecnologie su misura per le loro esigenze. Questo progetto di tesi vuole esplorare l'integrazione di tecnologie basate sul linguaggio, socially assistive robots (SARs) e IoT al fine di supportare gli anziani nella vita quotidiana a casa o in un centro di assistenza. Nello specifico, è stata sviluppata un'interfaccia vocale per robot sociali all'interno di smart spaces di tipo domestico che consente ai robot di: - coinvolgere gli anziani in conversazioni, che integrino informazioni sull'ambiente circostante, relative ad attività quotidiane o a comportamenti coerenti con le condizioni ambientali interne ed esterne - agire da tramite per il controllo domotico La soluzione tecnologica è stata progettata per essere il più possibile indipendente dalla piattaforma robotica utilizzata e compatibile con diversi sistemi robotici. Sono stati condotti due studi sperimentali su piccola scala (5 partecipanti ciascuno), il primo mirato a comprendere la complessità percepita dell'utilizzo di un sistema vocale in ambienti domestici e il secondo incentrato sull'impatto che la forma e il movimento del sistema hanno sull'interazione uomo-computer in questi ambienti . Per quanto riguarda il primo studio, i risultati dei questionari NASA-TLX rivelano che il sistema dovrebbe essere reso più veloce e il riconoscimento vocale dovrebbe essere migliorato. I risultati del secondo studio evidenziano come sistemi dotati di corpo e in grado di muoversi abbiano maggiore engagement e presenza sociale rispetto a sistemi statici.
Socially assistive robot in smart spaces for ambient assisted living
Toscano, Eleonora
2020/2021
Abstract
Considering that our society is aging and the number of elderly people is increasing every day, it is an ethical imperative to develop technologies tailored to their needs. The context of this thesis was to explore the integration of speech-based technologies, socially assistive robots (SARs), and IoT to support the elderly in everyday life at home or care center. Specifically, a speech-based interface has been developed for social robots connected with smart space sensors and actuators, enabling robots to: - engage the elderly in context-aware conversations related to everyday tasks or to behaviors that are consistent with indoor and outdoor environmental conditions - to act as a proxy for domotic control The technological solution has been designed to be as much as possible robot independent and compatible with different robotic systems. We ran two small scale experimental studies (5 participants each), the first aimed at understanding the perceived complexity of using a speech-based system in domestic environments and the second focused on the impact embodiment and movement have on human-computer interaction in said environments. Concerning the first study, results of NASA-TLX questionnaires reveal that the system should be sped up and speech recognition should be improved. Results of the second study highlight the increased engagement and social presence of embodied system able to move, compared to static ones.| File | Dimensione | Formato | |
|---|---|---|---|
|
2021_10_Toscano.pdf
accessibile in internet per tutti
Descrizione: Thesis
Dimensione
23.83 MB
Formato
Adobe PDF
|
23.83 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/179940