Monitoring cognitive performance during virtual flight training sessions is important to prevent negative experiences and ensure proper training. Due to the demanding nature of the task and reluctance of student pilots to report their cognitive state, we propose using Long Short-Term Memory Autoencoder networks for anomaly detection to monitor cognitive decline. This application adds automatic processing capabilities to the flight simulator, and the resulting information can be used by flight instructors for enhanced debriefing.
Unsupervised cognitive monitoring in a mixed-reality flight simulator for smart debriefing
01.06.2023
1125616 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch