疲勞度之即時偵測與分析

Real-time Detection and Analysis of Fatigue Level

曾國剛(Guo-Gang Zeng)

指導教授 ： 譚旦旭 ； 黃文增

國立臺北科技大學/電資學院/電機工程系研究所/碩士(2010年)

摘要

現代人由於工作繁忙與壓力大，且缺乏運動及適當休息；因精神疲勞而導致的行車意外經常發生。因此，本文研究與設計一系統，應用光體積變化描記圖原理製作的感測器量測駕駛心率訊號，經由ZigBee傳輸機制傳送至個人電腦，將訊號以Berger提出的方法進行重新取樣後，再經由快速傅立葉演算法將訊號從時域轉換至頻域，並依歐洲心臟醫學會及北美心律電生理學會對心率變異各頻帶定義之標準，取出交感/副交感神經平衡指標做為評估駕駛精神疲勞的指數。實驗對象為30名正常男性，年齡在23歲左右，隨機分成A(早上)、B(下午)、C(晚上)三組，各組中每位受測人分別模擬駕駛，並於駕駛90分鐘後休息30分鐘。實驗期間每隔10分鐘分析一次生理訊號，並要求受測者填寫自我疲勞評估表格。實驗結果顯示，隨駕駛任務的進行，受測者自我評估之疲勞程度逐漸上升，但其平衡指標的變化趨勢並非是持續上升或下降的，所以再利用模糊理論來判斷駕駛當前之疲勞狀態。白天時，當系統判斷駕駛當前疲勞狀態(共5個等級)在等級4以上或是第二次偵測到在等級3以上時，即警告駕駛休息；晚上時，當判斷駕駛當前之疲勞狀態在等級3以上，即提出警告建議駕駛休息，以降低因駕駛精神疲勞而導致的遺憾事件發生。實驗結果證明，本系統能有效地在駕駛者精神疲勞時對其發出警告。

關鍵字

精神疲勞 ； 心率 ； 交感/副交感神經平衡指標 ； 模糊理論

並列摘要

Nowadays, there are many car accidents resulting from fatigue driving. The reasons that cause fatigue driving include heavy workload, pressures, lack of exercise and regular rest. Therefore, this study aims to design and implement a system to avoid car accidents due to mental fatigue driving.
The proposed system is developed based on the photoplethysmography scheme to detect the driver’s heart rate data, where ZigBee is used to wirelessly transmit data to personal computer. Next, the data is resampled by Berger algorithm and then is transformed from time domain to frequency domain by using fast Fourier transform (FFT) for successive analysis. The index of the sympathovagal balance for assessing driving mental fatigue is derived by following related standards of the European Society of Cardiology and the North American Society of Pacing.
A series of experiments is conducted with 30 young males whose ages are between 22 and 24 year-old, randomly assigned to 3 groups, which are group A driving in morning period, group B driving in afternoon period, and group C driving in evening period. Each participant drives for 90 minutes then takes a rest of 30 minutes. In the experiment, each participant receives bio-signal analysis and detection every 10 minutes and fills a form for fatigue assessment. The experimental result indicates that the fatigue level increases with driving time. However, the tendency of sympathovagal balance is not always linear. Therefore, this study further employs the fuzzy theory to evaluate the fatigue level of participant.
During the day time, when the fatigue level (including 5 levels) of the driver is higher than 4 or has been detected twice higher than 3, the system will alert the driver and suggest to take a rest. On the other hand, in the evening, when the fatigue level of the driver is higher than 3, the system will alert the driver and suggest to take a rest, so as to reduce the possibility of accident caused by mental fatigue driving. The results of experiments demonstrate that the system can effectively warn the driver when his/her fatigue level is higher than the threshold; thus verifying the effectiveness of the proposed system for practical application.

並列關鍵字

Mental Fatigue ； Heart Rate ； Sympathovagal Balance ； Fuzzy Theory

參考文獻

[21] 張肇峰，“利用PPG分析自然心律變異”，碩士論文，中原大學，2008。

[34] 紀進鴻，“四通道生理訊號儲存與回讀系統”，碩士論文，中原大學，2001。

[22] 胡志彰，“以血氧濃度為基礎的肌肉疲勞評估系統初探”， 碩士論文，成功大學，2007。

[4] 姚景升，“同步量測心電圖和脈波儀器之設計製作和心率變異度與血壓變異度之分析”，中原大學醫學工程學系，碩士學位論文，2002。

[6] 黃仲麟，自行車玩家生醫資訊守護神，國立臺北科技大學電腦與通訊研究所碩士論文，2009。