지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 발행연도
- 2018.12
- 수록면
- 681 - 689 (9page)
- DOI
- 10.5143/JESK.2018.37.6.681
이용수
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
Objective: The purpose of this study is to analyze characteristics of flight performance and heart rate variability features and develop cognitive workload classification model in order to detect cognitive workload levels of pilots.
Background: Pilots can experience high cognitive workload conditions by performing various tasks simultaneously during air operation mission. Accurate detection of the cognitive workload state is required to improve the survivability of pilots and ensure flight safety.
Method: Five student pilots in 20s performed dual task consisting of simulated flight task with baseline task and three levels of N-back task (0-back, 1-back, 2-back). Three flight performance features (standard deviation of altitude, heading, airspeed) and two heart rate variability features (standard deviation of NN (SDNN), Low Frequency to High Frequency ratio (LF/HF)) were measured from the flight simulation experiment with photoplethysmogram (PPG) sensor. Cognitive workload classification model was developed by pre-processing input features and output classes, applying five classifiers (Decision tree (DT), Support vector machine (SVM), K-nearest neighbor (KNN), Bayesian, Ensemble), and validating the model performance.
Results: High cognitive workload level (2-back) was 327%, 161%, and 1.2% higher than baseline in terms of standard deviation of altitude, standard deviation of airspeed, and LF/HF respectively. DT, SVM, KNN, Bayesian, and Ensemble classifier performance was 94.4%, 91.1%, 96.6%, 93.3%, and 96.6% with two output classes, 62%, 51.4%, 64.6%, 57.1%, and 64.6% with three output classes, 36.3%, 36%, 48%, 36.3%, and 49% with four output classes respectively.
Conclusion: The best classifier performance for detecting cognitive workload levels of pilots was acquired from an ensemble classifier with two output classes (baseline, 2-back) with an accuracy of 96.6%.
Application: The cognitive workload classification model using the ensemble classifier in this study can contribute to the development of a system capable of providing warning signal in real time under pilot"s cognitive overload situation.
Background: Pilots can experience high cognitive workload conditions by performing various tasks simultaneously during air operation mission. Accurate detection of the cognitive workload state is required to improve the survivability of pilots and ensure flight safety.
Method: Five student pilots in 20s performed dual task consisting of simulated flight task with baseline task and three levels of N-back task (0-back, 1-back, 2-back). Three flight performance features (standard deviation of altitude, heading, airspeed) and two heart rate variability features (standard deviation of NN (SDNN), Low Frequency to High Frequency ratio (LF/HF)) were measured from the flight simulation experiment with photoplethysmogram (PPG) sensor. Cognitive workload classification model was developed by pre-processing input features and output classes, applying five classifiers (Decision tree (DT), Support vector machine (SVM), K-nearest neighbor (KNN), Bayesian, Ensemble), and validating the model performance.
Results: High cognitive workload level (2-back) was 327%, 161%, and 1.2% higher than baseline in terms of standard deviation of altitude, standard deviation of airspeed, and LF/HF respectively. DT, SVM, KNN, Bayesian, and Ensemble classifier performance was 94.4%, 91.1%, 96.6%, 93.3%, and 96.6% with two output classes, 62%, 51.4%, 64.6%, 57.1%, and 64.6% with three output classes, 36.3%, 36%, 48%, 36.3%, and 49% with four output classes respectively.
Conclusion: The best classifier performance for detecting cognitive workload levels of pilots was acquired from an ensemble classifier with two output classes (baseline, 2-back) with an accuracy of 96.6%.
Application: The cognitive workload classification model using the ensemble classifier in this study can contribute to the development of a system capable of providing warning signal in real time under pilot"s cognitive overload situation.
목차
1. Introduction
2. Method
3. Results
4. Discussion
References
참고문헌 (15)
참고문헌 신청
[학술대회논문] - Coyne, J. T. / 2009 / Applying Real Time Physiological Measures of Cognitive Load to Improve Training / Proceedings of HCI International Conference on Foundations of Augmented Cognition: Neuroergonomics and Operational Neuroscience : 469 ~ 478
[학술대회논문] - Dietterich, T.H / 2000 / Ensemble Methods in Machine Learning / Proceedings of First International Workshop on Multiple Classifier Systems
[보고서] - Gawron, V.J. / 2002 / Airplane Upset Training Evaluation Report
[학술지(정기간행물)] - Katisis, C.D. / 2008 / Toward Emotion Recognition in Car-Racing Drivers: A Biosignal Processing Approach / IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans 38 (3) : 502 ~ 512
[단행본] - Kim, B.S. / 2015 / Application and Interpretation of HRV in Stress Clinic / Panmun Education
[학술대회논문] - Dietterich, T.H / 2000 / Ensemble Methods in Machine Learning / Proceedings of First International Workshop on Multiple Classifier Systems
[보고서] - Gawron, V.J. / 2002 / Airplane Upset Training Evaluation Report
[학술지(정기간행물)] - Katisis, C.D. / 2008 / Toward Emotion Recognition in Car-Racing Drivers: A Biosignal Processing Approach / IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans 38 (3) : 502 ~ 512
[단행본] - Kim, B.S. / 2015 / Application and Interpretation of HRV in Stress Clinic / Panmun Education
함께 읽어보면 좋을 논문
논문 유사도에 따라 DBpia 가 추천하는 논문입니다. 함께 보면 좋을 연관 논문을 확인해보세요!
-
Prediction and Classification of Pilot’s Cognitive Workload via Deep Learning of Multivariate Data
대한인간공학회 학술대회논문집
2019 .10
-
Error Analysis of Multidimensional Evaluation of Pilot’s Workload Based on Objective Measures
제어로봇시스템학회 국제학술대회 논문집
2015 .10
-
Development of Real-time Assessment System of Driver’s Cognitive Workload Using Artificial Neural Network on ECG
대한인간공학회 학술대회논문집
2017 .04
-
신체적 작업 부하가 시간 인지에 미치는 영향
대한인간공학회지
2022 .10
-
원전 기동운전에서 운전원 작업부하 주관적평가
대한인간공학회 학술대회논문집
2020 .10
-
Classification of Driver’s Cognitive Workload Levels using Artificial Neural Network on ECG
대한산업공학회 추계학술대회 논문집
2015 .11
-
Characteristics of the EEG information about driving workload on the winding road
대한교통학회 학술대회지
2016 .09
-
Comparison Study of Potential Workload Index in a Simulated Multiple-UAV Operation Environment
제어로봇시스템학회 국제학술대회 논문집
2017 .10
-
수중 표적 식별을 위한 앙상블 학습
멀티미디어학회논문지
2015 .11
-
Effects of concurrent cognitive tasks on the performance and stress during a physical task
대한인간공학회 학술대회논문집
2020 .10
-
자율주행 상황에서 제어권 전환 요청 시기가 운전자 작업부하에 미치는 영향
대한인간공학회지
2019 .04
-
인지 부하 정도를 추정하기 위한 뇌파 지표 검증
대한전기학회 학술대회 논문집
2019 .02
-
인지적 작업부하에 따른 경추부 근육 활성도 정량화에 관한 연구
대한인간공학회 학술대회논문집
2022 .10
-
A Preliminary Result of Systematic Literature Review on Human Status Detection using Machine Learning Techniques
대한인간공학회 학술대회논문집
2021 .11
-
Comparison of Workload Perception for Original and Modified Cognitive Tasks
Industrial Engineering & Management Systems
2019 .12
-
특허키워드분석을 위한 베이지안 회귀모형
한국컴퓨터정보학회논문지
2016 .01
-
Pilot’s Cognitive State Recognition using Wavelet Singular Entropy and Gaussian Process Classification via Full Flight Simulation
제어로봇시스템학회 국제학술대회 논문집
2015 .10
-
굴착기 내구시험 모니터링을 위한 작업부하 지표 개발
드라이브·컨트롤
2022 .12
-
Study on the Relationship between Adolescents" Self-esteem and their Sociality
한국컴퓨터정보학회논문지
2016 .01
-
국방품질보증 업무량과 중요성 비교를 통한 발전방향 연구
신뢰성응용연구
2023 .06
이 논문의 저자 정보
최근 본 자료
댓글(0)
0