Emotion Analysis Architecture Based on Face and Physiological Sensing Applied with Flight Simulator

Abstract

This work presents an architecture as an important contribution regarding to emotional events along tasks based on flight simulations. This architecture considered eight beginner users of flight simulator (n = 8) while they execute a simulated flight according with the basic concepts of visual flight rules (VFR). The acquired physiological sensing were: heart rate (HR), electroencephalography (EEG) and galvanic skin response (GSR). One small camera was also used to record the users’ face in order to extract, after the post-processing, the emotions of the user during the flight. The considered emotions were: happy, sad, angry, surprised, scared and disgust. Initial analysis of the GSR signals shown that the takeoff task presented 13% more variability (or emotional events) between the climb and approach tasks together; and in the same way, the landing task presented 16% more variability between the climb and approach tasks together, what shows the importance of these researches in flight safety context, mainly in these critical phases.