Xinyun Hu
Professor Gabriel Lodewijks

Operator fatigue is a significant cause of operational errors in aviation and ground transportation, posing a serious threat to transportation safety. A wide range of non-invasive fatigue detection methods including a number of objective eye-tracking metrics have been explored in the transportation field. The development of eye movement-based fatigue detection technology for use in real world application has been hampered by discrepancies between the results of existing studies. My research aim is to address this issue by exploring which eye metrics most effectively indicate operator fatigue, with the intention of assisting fatigue detection in air and ground transport. In this study, eye movement data were collected while operators attended to their tasks. Features of their eye movements were then analysed to explore correlations with their subjective state to identify which features best indicate fatigue. Successfully identifying eye movement-related fatigue indicators could identify the presence of fatigue in transport operator, which could then be used to prompt the operator to take a rest or even inform a third party that intervention is necessary. This could considerably enhance transport safety and reduce accidents caused by fatigue and human error.