Mental workload recognition of carrier-based aircraft pilots based on portable photoplethysmography equipment
ZHU Wenbing
ZHANG Chenyang
YUAN Jiajun
XU Fang
MA Yuan
JIANG Chaozhe
Abstract：Objective To identify the mental workload of pilots in different flight stages by collecting physiological data of pilots based on Su-33 flight simulator platform.Methods Based on aircraft carrier-based flight tasks,the data of National Aeronautics and Space Administration-Task Load Index scale and photoplethysmography of 6 pilot cadets were obtained in 3 flight stages(takeoff and climb,cruise,and approach and landing).Random forest(RF)was used to evaluate the importance of heart rate variability(HRV)indexes,and statistical analysis was used to compare differences across flight stages.The HRV indexes that best reflected mental workload were selected.The three machine learning algorithms[support vector machine,K-nearest neighbor(KNN)algorithm,and RF],as well as two neural network algorithms(convolutional neural network and bidirectional long short-term memory network)were used for mental workload recognition.Results The mental workload levels were the highest during the approach and landing phase,followed by the cruise phase,and the lowest during the takeoff and climb phase.Each classification model had a good recognition effect on mental workload,and the data collected based on portable photoplethysmography equipment had a high accuracy,with KNN algorithm performing the best,achieving an accuracy rate of 92.9％.Conclusion The HRV indexes,combined with a variety of machine learning and neural network algorithms,can effectively identify mental workload levels across different flight phases,offering a scientific basis for pilot training and cognitive evaluation.
Keywords：aviation safetypilotsmental workloadmachine learningneural networkportable photoplethysmography monitoring equipment
Publication Date：2025-01-27
Online Publishing Date：2026-08-26(First online date of this platform, not the publication date of the document)