在偵測疲勞的實驗中，目前最主流的方法是使用部分睡眠剝奪或是完全睡眠剝奪。本論文提出使用模擬駕駛來替代睡眠剝奪的過程，透過讓受試者在這段期間集中於耗費精神的事項，以縮短受試者進入疲勞狀態的時間，並分析在這種情況下各個疲勞分類標準之間的優劣。

在本文中使用精神運動警覺性任務(Psychomotor Vigilance Test)以及卡羅林斯卡嗜睡量表(Karolinska sleepiness scale)來測量受試者的狀態，並且在駕駛過程與PVT測試中同時量測心電圖(Electrocardiography)訊號。為了從PVT測試中提取特徵值，使用了漂移擴散模型(Drift-Diffusion model)，將PVT測試的特徵值提取成漂移率(Drift rate)與非決定時間(Non-decision time)。使用了KSS、失誤次數、ΔRT>250ms次數、漂移率、非決定時間作為疲勞分類參數。將上述參數的資料分成兩組，並使用魏克生等級和檢定(Wilcoxon rank sum test)來尋找疲勞閾值。用上述閥值將ECG訊號的分類後作為支持向量機(support vector machine)的輸入，分析各項疲勞分類標準的差異。

我們計算出的疲勞分類閾值與其餘使用睡眠剝奪的實驗做比對後十分相似，證明了不使用睡眠剝奪的實驗流程確實可以使受試者進入疲勞狀態。其中非決策時間、失誤次數與漂移率的準確率分別為95.83%、94.64%與91.27%，顯示出DDM的模擬參數是優於單純客觀測量與主觀測量。

In fatigue test, the most frequently used methods are total sleep deprivation and partial sleep deprivation. Therefore, this article proposes to replace the process of sleep deprivation with driving simulation in order to shorten the time to fatigue patients, and analyze the pros and cons between the various fatigue classification standards.

In this paper, we use electrocardiography (ECG) signal, psychomotor vigilance task (PVT) and Karolinska sleepiness scale (KSS) to measure the fatigue state of subjects. In order to extract fatigue features from the PVT test, we also use drift diffusion model (DDM) to extract the data of the PVT test into drift rate and non-decision time.

KSS, the number of lapse, the number of ΔRT>250ms, drift rate, and non-decision time are used as fatigue classification parameters. Divide the data of the above parameters into two groups, and use the Wilcoxon rank sum test to find the fatigue threshold. Use the above threshold to classify the ECG signal as the input of the support vector machine.

The fatigue classification threshold that we calculated is very similar to other experiments that use sleep deprivation. It proves that the experimental without sleep deprivation can make the subjects fatigue. The accuracy of non-decision time, number of lapse and drift rate are 95.83%, 94.64% and 91.27%. The parameters of DDM are better than objective measurement and subjective measurement.

摘要....................................................I
誌謝....................................................III
Abstract................................................V
第一章 緒論..............................................1
1.1 研究背景.............................................1
1.2 研究動機.............................................2
1.3 論文大綱.............................................3
第二章 文獻回顧...........................................5
2.1 疲勞測量方式..........................................5
2.1.1 車輛表現............................................5
2.1.2 駕駛行為............................................6
2.1.3 生理訊號............................................7
2.1.4 主觀評估............................................8
2.2 疲勞實驗方案..........................................8
2.2.1 完全睡眠剝奪(Total Sleep Deprivation)...............8
2.2.2 部分睡眠剝奪(Partial Sleep Deprivation).............9
2.2.3 不使用睡眠剝奪......................................10
2.3 疲勞分類指標及方法....................................10
2.3.1 精神運動警覺性任務(PVT).............................11
2.4 本章小結.............................................11
第三章 實驗設計...........................................13
3.1 實驗流程.............................................13
3.1.1 實驗器材...........................................14
3.1.2 實驗受試者.........................................15
3.2 實驗資料蒐集..........................................15
3.2.1 精神運動警覺性測試(PVT).............................15
3.2.2 心率變異性(HRV)....................................16
3.2.3 卡羅林斯卡嗜睡量表(KSS).............................17
3.3 本章小結.............................................18
第四章 資料處理與疲勞分類標準..............................19
4.1 ECG資料處理..........................................19
4.1.1 去除噪音...........................................20
4.1.2 R波檢測............................................21
4.1.3 HRV特徵計算........................................21
4.2 PVT資料處理..........................................22
4.2.1 失誤(Lapse)次數及其延伸.............................22
4.2.2 漂移擴散模型(DDM)...................................23
4.3 疲勞分類標籤..........................................25
4.3.1 卡羅林斯卡嗜睡量表(KSS)..............................27
4.3.2 精神運動警覺性任務(PVT)..............................29
4.3.3 漂移擴散模型(DDM)...................................30
4.4 支援向量機模型(SVM)....................................32
4.4.1 資料比例調整.........................................32
4.4.2 輸入特徵.............................................34
4.4.3 調整參數.............................................38
4.4.4 本章小結.............................................39
第五章 實驗結果與分析.......................................41
5.1 疲勞效果比較...........................................41
5.2 各項疲勞標籤的比較......................................42
5.3 真實道路駕駛............................................45
5.4 討論...................................................48
5.4.1 實驗流程.............................................48
5.4.2 模擬駕駛.............................................48
5.4.3 真實道路駕駛..........................................50
5.4.4 疲勞載體.............................................50
5.4.5 資料比例.............................................50
第六章 結論與未來發展.......................................53
6.1 結論...................................................53
6.2 未來應用...............................................54
6.3 未來工作...............................................54
參考文獻...................................................55

[1]A. W. MacLean, D. R. Davies, and K. Thiele, “The hazards and prevention of driving while sleepy,” Sleep Medicine Reviews, vol. 7, no. 6, pp. 507–521, 2003.

[2]H. Katherine and S. Jennifer, “Saving lives through advanced vehicle safety technology : intelligent vehicle initiative,” pp. 1–12, 2005.

[3]E. Grandjean, “Fatigue in industry,” British journal of industrial medicine , vol. 36, no. 3, pp. 175–186, 1979.

[4]D. A. Grant, K. A. Honn, M. E. Layton, S. M. Riedy, and H. P. Van Dongen, “3-minute smartphone-based and tablet-based psychomotor vigilance tests for the assessment of reduced alertness due to sleep deprivation,” Behavior Research Methods, vol. 49, no. 3, p.1020–1029, 2016.

[5]A. Kosmadopoulos, C. Sargent, X. Zhou, D. Darwent, R. W. Matthews, D. Dawson, and G. D. Roach, “The efficacy of objective and subjective predictors of driving performance during sleep restriction and circadian misalignment,” Accident Analysisamp; Prevention, vol. 9, p. 445–451, 2017.

[6]N. Theresia Br. Pasaribu, R. ., A. Prijono, R. Pramono Adhie, W. Halim, and R. Maini Heryanto, “Performance male and female drivers in drowsiness system based on psychomotor vigilance task test,” International Journal of Engineering amp; Technology, vol. 7, no. 2.13, p. 421, 2018.

[7]C. D. R. (EU), “Drowsiness and attention warning systems and amending ...2020,” 2020, https://members.wto.org/crnattachments/2021/TBT/EEC/21_0621_01_e.pdf.

[8]J. Verster and T. Roth, “Standard operation procedures for conducting the on-the-road driving test, and measurement of the standard deviation of lateral position (SDLP),”International journal of general medicine, vol. 4, pp. 359–71, May. 2011.

[9]R. Senaratne, D. Hardy, B. Vanderaa, and S. Halgamuge, “Driver fatigue detection by fusing multiple cues,” Advances in Neural Networks–ISNN 2007, p. 801–809.

[10]Z. Mu, J. Hu, and J. Min, “Driver fatigue detection system using electroencephalography signals based on combined entropy features,” Applied Sciences, vol. 7, no. 2, p. 150, 2017.

[11]A. Chowdhury, R. Shankaran, M. Kavakli, and M. M. Haque, “Sensor applications and physiological features in drivers' drowsiness detection: A review,” IEEE Sensors Journal, vol. 18, no. 8, pp. 3055–3067, 2018.

[12]T. C. Chieh, M. M. Mustafa, A. Hussain, S. F. Hendi, and B. Y. Majlis, “Development of vehicle driver drowsiness detection system using electrooculogram (EOG),” 1st International Conference on Computers, Communications, Signal Processing with Special Track on Biomedical Engineering, pp. 165–168, 2005.

[13]A. A. Hayawi and J. Waleed, “Driver’s drowsiness monitoring and alarming auto-system based on EOG signals,” 2nd International Conference on Engineering Technology and its Applications (IICETA), pp. 214–218, 2019.

[14]F. Abtahi, A. Anund, C. Fors, F. Seoane, and K. Lindecrantz, “Association of drivers’ sleepiness with heart rate variability: A pilot study with drivers on real roads,” EMBEC & NBC 2017, pp. 149–152, 2018.

[15]R. Bhardwaj, P. Natrajan, and V. Balasubramanian, “Study to determine the effectiveness of deep learning classifiers for ECG based driver fatigue classification,” IEEE 13th International Conference on Industrial and Information Systems (ICIIS), pp. 98–102, 2018.

[16]M. Ingre, T. Åkerstedt, B. Peters, A. Anund, and G. Kecklund, “Subjective sleepiness, simulated driving performance and blink duration: examining individual differences,” Journal of Sleep Research, vol. 15, no. 1, pp. 47–53, 2006. [Online]. Available: https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2869.2006.00504.x

[17]N. Theresia Br. Pasaribu, R. ., A. Prijono, R. Pramono Adhie, W. Halim, and R. Maini Heryanto, “Performance male and female drivers in drowsiness system based on psychomotor vigilance task test,” International Journal of Engineering & Technology, vol. 7, no. 2.13, p. 421, 2018

[18]M. Basner and D. Dinges, “An Adaptive-Duration Version of the PVT Accurately Tracks Changes in Psychomotor Vigilance Induced by Sleep Restriction,” Sleep, vol. 35, no. 2,pp. 193–202, 02 2012. [Online]. Available: https://doi.org/10.5665/sleep.1620

[19]BITalino (r)evolution Plugged Kit Data Sheet, BITalino, 2016, rev. A.

[20]A. Wilson, J. Dollman, K. Lushington, and T. Olds, “Reliability of the 5-min psychomotor vigilance task in a primary school classroom setting,” Behavior Research Methods, vol.42,no. 3, p. 754–758, 2010.

[21]S. Loh, N. Lamond, J. Dorrian, G. Roach, and D. Dawson, “The validity of psychomotor vigilance tasks of less than 10-minute duration,” Behavior Research Methods, Instruments, & Computers, vol. 36, no. 2, pp. 339–346, 2004. [Online]. Available: https://doi.org/10.3758/BF03195580

[22]M. Basner and D. F. Dinges, “Maximizing sensitivity of the psychomotor vigilance test (PVT) to sleep loss,” Sleep, vol. 34, no. 5, pp. 581–591, 05 2011. [Online]. Available: https://doi.org/10.1093/sleep/34.5.581

[23]F. Shaffer and J. P. Ginsberg, “An overview of heart rate variability metrics and norms,” Frontiers in Public Health, vol. 5, p. 258, 2017. [Online]. Available: https://www.frontiersin.org/article/10.3389/fpubh.2017.00258

[24]A. Miley, G. Kecklund, and T. Åkerstedt, “Comparing two versions of the karolinska sleepiness scale (KSS),” Sleep and Biological Rhythms, vol. 14, no. 3, pp. 257–260, 2016. [Online]. Available: https://doi.org/10.1007/s41105-016-0048-8

[25]M. Awais, N. Badruddin, and M. Drieberg, “A hybrid approach to detect driver drowsiness utilizing physiological signals to improve system performance and wearability,” Sensors, vol. 17, no. 9, 2017. [Online]. Available: https://www.mdpi.com/1424-8220/17/9/1991

[26]J.-Y. Hong, “A driver fatigue detection system based on heart rate variability,” Aug 2021. [Online]. Available: https://hdl.handle.net/11296/dp4yr6

[27]L. G. Tereshchenko and M. E. Josephson, “Frequency content and characteristics of ventricular conduction,” Journal of electrocardiology, pp. 933–937, 2015. [Online].
Available: https://pubmed.ncbi.nlm.nih.gov/26364232

[28]J. Pan and W. J. Tompkins, “A real-time QRS detection algorithm,” IEEE Transactionson Biomedical Engineering, vol. BME-32, no. 3, pp. 230–236, 1985.

[29]R. Buendia, F. Forcolin, J. Karlsson, B. A. Sjöqvist, A. Anund, and S. Candefjord, “Deriving heart rate variability indices from cardiac monitoring—an indicator of driver sleepiness,” Traffic Injury Prevention, vol. 20, no. 3, pp. 249–254, 2019. [Online]. Available: https://doi.org/10.1080/15389588.2018.1548766

[30]M. Malik, J. T. Bigger, A. J. Camm, R. E. Kleiger, A. Malliani, A. J. Moss, and P. J. Schwartz, “Heart rate variability: Standards of measurement, physiological interpretation, and clinical use,” European Heart Journal, vol. 17, no. 3, pp. 354–381, Mar 1996. [Online]. Available: https://doi.org/10.1093/oxfordjournals.eurheartj.a014868

[31]W. G. Bickley, “Piecewise Cubic Interpolation and Two-Point Boundary Problems,” The Computer Journal, vol. 11, no. 2, pp. 206–208, 08 1968. [Online]. Available: https://doi.org/10.1093/comjnl/11.2.206

[32]A. Patanaik, C. K. Kwoh, E. C. Chua, J. J. Gooley, and M. W. Chee, “Classifying vulnerability to sleep deprivation using baseline measures of psychomotor vigilance,” Sleep, vol. 38, no. 5, pp. 723–734, 05 2015. [Online]. Available: https://doi.org/10.5665/sleep.4664

[33]R. Ratcliff and G. McKoon, “The diffusion decision model: Theory and data for two-choice decision tasks,” Neural Computation, vol. 20, no. 4, p. 873–922, 2008.

[34]A. Patanaik, V. Zagorodnov, and C. K. Kwoh, “Parameter estimation and simulation for one-choice ratcliff diffusion model,” Proceedings of the 29th Annual ACM Symposium on Applied Computing, 2014.

[35]R. S. M. de Barros, J. I. G. Hidalgo, and D. R. de Lima Cabral, “Wilcoxon rank sum test drift detector,” Neurocomputing, vol. 275, pp. 1954–1963, 2018. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0925231217317022

[36]M. Basner, E. Hermosillo, J. Nasrini, S. McGuire, S. Saxena, T. M. Moore, R. C. Gur, and D. F. Dinges, “Repeated administration effects on psychomotor vigilance test performance,” Sleep, vol. 41, no. 1, 2017.