EEG-based stress evaluation in a ship’s bridge simulator based assessment

• المؤلف الرئيسي: Ley, Daryl Jun Rong
• مؤلفون آخرون: Ang Hock Eng
• التنسيق: Final Year Project
• اللغة: English
• منشور في: 2017
• الموضوعات: DRNTU::Engineering::Mechanical engineering::Assistive technology
• الوصول للمادة أونلاين: http://hdl.handle.net/10356/70698
• المؤسسة: Nanyang Technological University

Human factor is one of the repeatedly cited source for causing maritime accidents. Past researches and studies made use of different methodologies to study human factor and using of bio-signals reflected great accuracies and results. With today’s technological advancement, such technologies are readily available and cost efficient. As such, the purpose of this study is to advance the study and research in the field of human factor studies in maritime domain by using EEG brain state monitoring technology. This study is a collaborative effort between Nanyang Technological University (NTU), Fraunhofer IDM @ NTU and Maritime Institute @ Singapore Polytechnic (MI@SP) which aims to develop a novel method of research for EEG stress recognition techniques in a ship’s bridge simulator test setting. In this study, 18 subjects participated in 4 bridge simulation exercises which are supposed to induce varying levels of stress. The EEG data are recorded for the exercises using the Emotiv EPOC headsets. Support Machine Vector (SVM) classifier is then used for 2-level emotion and 4-level workload recognition. 8 possible stress levels are then derived from the recognised emotion and workload data. With synchronised video footage and the EEG data, the stress levels estimated were studied using graphical second-by-second analyses and statistical analyses. The results from this study found that there are indeed correlations between EEG data and the demanding situations occurring in the bridge simulations. Demanding and difficult situations can be identified from the peaks in the graphical analyses. This study drew conclusive results for the effectiveness and validation of using EEG in maritime human factor studies. However, the statistical analyses did not show any significant findings.