@phdthesis{,
    author = {Masters, Matthew },
    title = {Real-Time Pilot Mental Workload Prediction Through the Fusion of Psychophysiological Signals},
    editor = {},
    booktitle = {},
    series = {},
    journal = {},
    address = {},
    publisher = {},
    edition = {},
    year = {2023},
    isbn = {},
    volume = {},
    number =  {},
    pages = {},
    url = {},
    doi = {},
    keywords = {Mental Workload, Psychophysiological, Physiological, Signal Processing, Pilot},
    abstract = {Military helicopter pilots and their aircraft form a unique system relied upon to be highly functioning. This work explores an aspect of the system largely ignored by the industrial developers of these systems, namely the mental workload of the pilots during flight. Supported by a systematic review of previously-published works, it is reasoned that mental workload is experienced uniquely by each individual and that it cannot be deduced through an analysis of the task load alone. A technical solution is developed and tested for predicting pilot mental workload in real-time which processes and fuses psychophysiological data from multiple sources supporting a multi-modal assessment. Specifically, signals processed include functional near-infrared spectroscopy (fNIRS), electrocardiography (ECG), electrodermal activity (EDA), respiration, and eye-movement-related signals. The unique signal processing chains are presented including the algorithms for extracting workload-relevant features and methods implemented to ensure robust data acquisition and processing. Experimentation of the system with ten operational military helicopter pilots and ten university students shows a moderate linear correlation between subjective and predicated mental workload (average Pearson’s correlation coefficient of 0.36±0.21). The individual feature with the strongest linear correlation to subjective mental workload is an instantaneous standard deviation of all deoxygenated hemoglobin channels recorded from the pre-frontal cortex. This discovery is significant as this feature has not been identified by previously-published works as being sensitive to mental workload. The developed system (including an in-cockpit display) demonstrates a high level of transparency required for effective human-machine systems. At last, a gauge in the cockpit for the most important sub-system in the human-machine team - the human!},
    note = {},
    
    school = {Universität der Bundeswehr München},
    
}