Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.818016
Title: Modelling, control and design of autonomous artificial avatars in human motor coordination task
Author: Lombardi, Maria
ISNI:       0000 0004 9359 0467
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Human coordination is a phenomenon that takes place in numerous daily activities, such as simple oral communication, walking in a crowd, clapping within an audience and when performing more complex coordinated activities such as playing in team sports or in musical ensembles. Unveiling the mechanisms that lead people to coordinate, adjust their movements properly, reach and maintain a stable coordinated behaviour represents a key challenge, both from a psychological and from a control point of view. Addressing this challenge is crucial, for example, to control artificial cyber-agents able to interact with people to perform common joint tasks. This thesis is concerned with the problem of designing an autonomous artificial agent able to move in a natural way in coordination with one or more humans. This is particularly relevant in the context of healthcare applications. Indeed, the use has been proposed of artificial agents coordinating their movements with those of patients suffering from social or motor disorders. Specifically, it has been shown that an artificial agent moving its end-effector with certain human kinematic properties could provide innovative and efficient rehabilitation strategies. In this thesis, human behaviour is studied through a simple yet effective coordination paradigm, where participants are asked to synchronise their hand motion. Keeping the same motor task, artificial agents with different control strategies are designed to interact with human participants so as to produce coordinated motion in different configurations. Different control approaches including those based on reinforcement learning are explored and validated via numerical simulations and experiments confirming the effectiveness of the proposed control architectures. The results of some additional work on the implementation of an exergame for motor rehabilitation of patient after stroke is also reported together with the analysis of leadership emergence in walking groups.
Supervisor: Marucci, Lucia Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.818016  DOI: Not available
Share: