Team, Visitors, External Collaborators
Overall Objectives
Research Program
Application Domains
New Software and Platforms
New Results
Bilateral Contracts and Grants with Industry
Partnerships and Cooperations
XML PDF e-pub
PDF e-Pub

Section: New Results

Self-affinity of an Aircraft Pilot's Gaze Direction as a Marker of Visual Tunneling

Abstract : For the last few years, a great deal of interest has been paid to crew monitoring systems in order to tackle potential safety problems during a flight. They aim at detecting any degraded physiological and/or cognitive state of an aircraft pilot, such as attentional tunneling or excessive focalization. Indeed, they might have a negative impact on his performance to pursue his mission with adequate flight safety levels. One of the usual approaches consists in using sensors to collect physiological signals which are analyzed in real-time. Two main families exist to process the signals. The first one combines feature extraction and machine learning whereas the second is based on deep-learning approaches but may require a large amount of labelled data. Here, we focused on the first family. In this case, various features can be deduced from the data by different approaches: spectrum analysis, a priori modelling and nonlinear dynamical system analysis techniques including the estimation of the self-affinity of the signals. In this paper, our purpose was to analyze whether the self-affinity of the pilot gaze direction can be related to his cognitive state. To this end, an experiment was carried out on 18 subjects in a representative aircraft environment based on a modified version of the software MATB-II. The scenarii were designed to elicit different levels of mental workload eventually associated to attentional tunneling. A database to train the machine learning step was first created by recording the directions of gaze of the subjects with an eye-tracker. The self-affinities of these signals were extracted with the Detrended Fluctuation Analysis method. They constituted the inputs of the classifier based on a Support Vector Machine. Then, new signals were analyzed and classified. Preliminary results showed promising abilities to detect attentional tunnelling episodes for different levels of mental workload.

Authors : Bastien Berthelot, Patrick Mazoyer, Sarah Egea, Jean-Marc André, Eric Grivel