Estimation de l'état fonctionnel de l'opérateur

Authors: Gagnon, Olivier
Advisor: Parizeau, Marc
Abstract: The assessment of an operator’s functional state (i.e., the multidimensional pattern of human psycho-physiological conditions that mediates performance) has great potential for increasing safety and reliability of critical systems. Machine learning, which has had success in recent years, is a technique which should be investigated for this task. An open question in the use of machine learning algorithms for the assessment of the operator’s functional state is the formalization of the operator’s state in an objective measure that can provide a training signal for the algorithms. This Master’s thesis introduces the decontextualized dynamic performance, a measure which enables the use of machine learning for many experimental tasks and many participants simultaneously.This work also explores the performances obtained by machine learning techniques in some contexts. The generalization of the trained models to new participants, or new tasks as well as the utilization of the training context is investigated.
Document Type: Mémoire de maîtrise
Issue Date: 2017
Open Access Date: 24 April 2018
Permalink: http://hdl.handle.net/20.500.11794/27618
Grantor: Université Laval
Collection:Thèses et mémoires

Files in this item:
SizeFormat 
33047.pdf3.23 MBAdobe PDFView/Open
All documents in CorpusUL are protected by Copyright Act of Canada.