Feasibility Research on Pilot Attention State Recognition Based on Eye Metrics

Release date:2022-04-29

Zhang YifanWang YuchaoZhang QinyuGe XianliangXu Jie 

Zhejiang UniversityHangzhou 310058China 

Abstract: The main tasks of the pilots in future highly automated aircraft will be tasks monitoring. One of the promising ways to promote performance and safety in such tasks is the development and utilization of real-time pilot attention recognition and intervention technologies in the cockpit. The purpose of the current study is to explore the feasibility of using eye metrics to identify the different states of sustained attention with the Gradual Onset Continuous Performance Task (GradCPT) paradigm. 30participants participated in the study and each completed three eightminute sessions of GradCPT tasks, while eye metrics are collected with an eye-tracker. "In the zone" and "out of the zone" periods during the task are identified according to the variance time course measure men. Eye-tracking data indicated that the mean and variation of pupil size, saccade duration, saccade peak velocity, and fixation duration measures are sensitive to the "in-the-zone" vs. "the out-of-the-zone" periods. The results suggest that it is feasible to recognize the attention state through eye metrics. Future research can use eye-tracking indicators as input features to build machine learning models of sustained attention for pilots. 

Key Words:human factors and ergonomics; human-machine interface; human factors in aviation; sustained attention; GradCPT

©2019 Chinese Aeronautical Establishment (CAE) All rights reserved. ICP:14045040