prikaz prve stranice dokumenta Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes
No public access
doctoral thesis
Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. urn:nbn:hr:168:583806
Petric, Frano
University of Zagreb
Faculty of Electrical Engineering and Computing
Department of Control and Computer Engineering

Institutional repository: FER Repository

Cite this document

Petric, F. (2018). Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes (Doctoral thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:583806

Petric, Frano. "Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes." Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:583806

Petric, Frano. "Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes." Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:583806

Petric, F. (2018). 'Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes', Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 26 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:583806

Petric F. Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes [Doctoral thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2018 [cited 2024 December 26] Available at: https://urn.nsk.hr/urn:nbn:hr:168:583806

F. Petric, "Robot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes", Doctoral thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2018. Available at: https://urn.nsk.hr/urn:nbn:hr:168:583806

Metadata
TitleRobot-assisted autism spectrum disorder diagnostics using partially observable Markov decision processes
Title (croatian)Robotom potpomognuta dijagnostika poremećaja iz autističnoga spektra primjenom djelomično osmotrivih Markovljevih procesa odlučivanja
AuthorFrano Petric
MentorZdenko Kovačić (mentor)
Committee memberZdenko Kovačić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
(Department of Control and Computer Engineering)
Zagreb
Defense date and country2018, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Universal decimal classification
(UDC)		621.3 - Electrical engineering
Abstract
The existing procedures for Autism Spectrum Disorder diagnosis are time consuming and challenging both for human evaluators and children being evaluated. The diagnosis of ASD relies solely on behavioral observations by experienced clinicians and occurrence of low agreement rates between different clinicians when evaluating a child suggests that there exists a need for a more objective approach to diagnostics and intervention. This thesis addresses that need by proposing a robot-assisted ASD diagnostic protocol consisting of four tasks adapted from ADOS. In this work the focus is on robot reasoning for ASD diagnostics. The main contribution of this thesis is a hierarchical Partially Observable Markov Decision Process framework that enables a humanoid robot to process the observations of child’s behavior, infer information about the unobservable state of the child and autonomously make decisions by selecting actions and tasks within the robot-assisted ASD diagnostic protocol. Each task of the protocol is modeled using a Mixed Observability Markov Decision Process model as a template. In order to formulate observation probabilities of task models, ASD experts are surveyed and their knowledge is encoded in the observation probabilities of task models. Expert knowledge also allowed for implementation of child behavioral models which are used to validate developed models. The model of the protocol is defined as a POMDP whose actions are tasks of the protocol. The interface between task and protocol models is formulated using regions of belief space of the task as observations for the protocol model. Following the successful validation through simulations with child behavioral models, task and protocol models are validated through experimental sessions with seven typically developing children and eight children with ASD. Results obtained through experiments show that the robot is capable of recognizing the behavior of the child and capable of differentiating different types of children, since the belief of the robot over the states of the child was comparable to assessment of autism experts.
Abstract (croatian)
Unatoč tome što postoje standardizirani testovi za dijagnostiku poremećaja iz autističnog spectra, dijagnostika predstavlja značajan izazov za stručno osposobljene ispitivače. Razlog tome je iznimna složenost dijagnostičkog postupka, koji se zbog nedostatka medicinskih markera oslanja isključivo na opažanje ponašanja djeteta. S obzirom na navedene problem, uočava se potreba za objektivnim mjerama socijalnih funkcija te se smatra da moderni roboti mogu biti korisni u istraživanju poremećaja iz autističnoga spektra. Roboti su već pronašli primjenu u intervencijskim postupcima za djecu s autizmom, ali se još uvijek ne koriste u dijagnostičkom postupku. Cilj je ove doktorske disertacije proširiti postojeće stanje u području korištenja robota za osobe s poremećajem iz autističnoga spektra postavljanjem temelja za autonomno izvođenje robotom potpomognutog protokola za dijagnostiku autizma. Glavni doprinos disertacije nova je hijerarhijska struktura upravljanja zasnovana na djelomično osmotrivim Markovljevim procesima (POMDP) odlučivanja koja omogućuje humanoidnom robotu obradu detekcija dječjeg ponašanja, zaključivanje o stanju djeteta te autonoman odabir akcija i zadataka dijagnostičkog protokola. Predložena struktura upravljanja sastoji se od modela dijagnostičkih zadataka koji su zasnovani na modelu Markovljevih procesa odlučivanja s mješovitom osmotrivošću (MOMDP). Modelima zadataka nadređen je model dijagnostičkog protokola koji je zasnovan na standardnom POMDP modelu. Sučelje između modela protokola i zadataka izvedeno je korištenjem regija stanja uvjerenosti robota na kraju zadataka kao opservacija za model protokola. Znanja stručnjaka s iskustvom u radu s djecom s autizmom zapisana su u obliku vjerojatnosti opservacija dječjeg socijalnog ponašanja u modelima dijagnostičkih zadataka. Saznanja o učestalosti određenih ponašanja kod djece korištena su za razvoj ponašajnih modela djeteta koji omogućuju simulaciju interakcije djeteta i robota s ciljem podešavanja i validacije razvijenih modela. Nakon provedenih simulacija, modeli zadataka i protokla eksperimentalno su provjereni provođenjem ispitivanja sa sedmero djece tipičnog razvoja i osmero djece s autizmom. Rezultati detaljne analize provedenih ispitivanje pokazuju da se stanje uvjerenosti robota na kraju protokola podudara s ocjenama stručnjaka za dijagnostiku autizma čime je utvrđeno da robot ima mogućnost prepoznavanja ponašanja djeteta te mogućnost razlikovanja djece iz različitih grupa.
Keywords
Keywords (english)
Languageenglish
URN:NBNurn:nbn:hr:168:583806
Study programmeTitle: Electrical Engineering and Computing
Study programme type: university
Study level: postgraduate
Academic / professional title: Doktor znanosti elektrotehnike i računarstva (Doktor znanosti elektrotehnike i računarstva)
Type of resourceText
Extent121 str. ; 30 cm.
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Created on2019-04-03 15:16:31