Date of Award

2023

Document Type

Dissertation

Degree Name

Ph.D.

Organizational Unit

Daniel Felix Ritchie School of Engineering and Computer Science, Electrical and Computer Engineering

First Advisor

Mohammad H. Mahoor

Second Advisor

Kimon Valavanis

Third Advisor

Timothy Sweeny

Fourth Advisor

Matthew Rutherford

Keywords

Artificial intelligence, Emotional intelligence, Social robotics

Abstract

Artificial Emotional Intelligence (AEI) bridges the gap between humans and machines by demonstrating empathy and affection towards each other. This is achieved by evaluating the emotional state of human users, adapting the machine’s behavior to them, and hence giving an appropriate response to those emotions. AEI is part of a larger field of studies called Affective Computing. Affective computing is the integration of artificial intelligence, psychology, robotics, biometrics, and many more fields of study. The main component in AEI and affective computing is emotion, and how we can utilize emotion to create a more natural and productive relationship between humans and machines.

An area in which AEI can be particularly beneficial is in building machines and robots for healthcare applications. Socially Assistive Robotics (SAR) is a subfield in robotics that aims at developing robots that can provide companionship to assist people with social interaction and companionship. For example, residents living in housing designed for older adults often feel lonely, isolated, and depressed; therefore, having social interaction and mental stimulation is critical to improve their well-being. Socially Assistive Robots are designed to address these needs by monitoring and improving the quality of life of patients with depression and dementia. Nevertheless, developing robots with AEI that understand users’ emotions and can reply to them naturally and effectively is in early infancy, and much more research needs to be carried out in this field.

This dissertation presents the results of my work in developing a social robot, called Ryan, equipped with AEI for effective and engaging dialogue with older adults with depression and dementia. Over the course of this research there has been three versions of Ryan. Each new version of Ryan is created using the lessons learned after conducting the studies presented in this dissertation. First, two human-robot-interaction studies were conducted showing validity of using a rear-projected robot to convey emotion and intent. Then, the feasibility of using Ryan to interact with older adults is studied. This study investigated the possible improvement of the quality of life of older adults. Ryan the Companionbot used in this project is a rear-projected lifelike conversational robot. Ryan is equipped with many features such as games, music, video, reminders, and general conversation. Ryan engages users in cognitive games and reminiscence activities. A pilot study was conducted with six older adults with early-stage dementia and/or depression living in a senior living facility. Each individual had 24/7 access to a Ryan in his/her room for a period of 4-6 weeks. The observations of these individuals, interviews with them and their caregivers, and analysis of their interactions during this period revealed that they established rapport with the robot and greatly valued and enjoyed having a companionbot in their room.

A multi-modal emotion recognition algorithm was developed as well as a multi-modal emotion expression system. These algorithms were then integrated into Ryan. To engage the subjects in a more empathic interaction with Ryan, a corpus of dialogues on different topics were created by English major students. An emotion recognition algorithm was designed and implemented and then integrated into the dialogue management system to empathize with users based on their perceived emotion. This study investigates the effects of this emotionally intelligent robot on older adults in the early stage of depression and dementia. The results of this study suggest that Ryan equipped with AEI is more engaging, likable, and attractive to users than Ryan without AEI. The long-term effect of the last version of Ryan (Ryan V3.0) was studied in a study involving 17 subjects from 5 different senior care facilities. The participants in this study experienced a general improvement in their cognitive and depression scores.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Hojjat Abdollahi

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

135 pgs

Discipline

Computer engineering, Electrical engineering



Share

COinS