Year: 2015
Type: dissertation
Abstract: This thesis presents the design of an assistive robotic system, developed to aid a growing elderly population in maintaining quality of life and age in place independently. The robot is designed to operate in a home environment and aid a person with the performance of independent activities of daily living, primarily focusing on guided meal preparation. This work begins with the development of a novel person-search approach using a time-discretized Hidden Markov Model (HMM). Subsequently the prototyping and development of the robotic system is discussed, followed by a benchmarking of the person-search system and an information gathering experiment evaluating the robot's interactive capabilities. Experiments show that the HMM-based person-search system is effective at locating a target person considerably faster than an uninformed baseline approach.
Source: TSpace (University of Toronto)
Author: Pavel Bovbel
Institution: University of Toronto
Language: English
Subfield: Social Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2570217236
