Study On Application Of Heart Rate Variability In Vehicle Driver Fatigue Detecting

Fatigue driving of vehicle drivers has been one of the factors that lead to traffic accidents. There're many methods to detect the driving fatigue, such as the way by the physiological or individual characteristics, and the way by the running status of vehicle. Among all these techniques, it's a noninvasive, convenient and rapid way to detect the fatigue state of vehicle driver using HRV (Heart Rate Variability) signals. The paper analyzed the time domain, frequency domain and non-linear specialities, and discussed the feasibility and application of the HRV indexes in the fatigue driving monitoring. The paper provided some new means and measures of the research on the fatigue driving monitoring of vehicle drivers.The paper collected 20 groups of electrocardiosignal during simulative driving firstly. Then the electrocardiosignal was resampled and passed the low-pass filter. After detecting R-wave using the threshold method, a high precision RR interval sequence, namely HRV signal was obtained. Summarizing and studying the postulate of time domain, frequency domain and non-linear analysis methods, and comparing the merits and faults of these three means, the individual parameters were chosen: (1) Time domain index: MEAN (average of RR interval), SDNN (standard deviation of RR interval), rMSSD (root-mean-square of difference-value of adjacent RR interval); (2) Frequency domain: LF (low-frequency power), HF (high-frequency power); (3) Non-linear index: D2 (correlation dimension), LE (Lyapunov exponent).The variation of each HRV signal index during driving was studied by the MATLAB software. Hereby it comes the conclusion as follows, along with the deepening of fatigue degree: (1) MEAN and rMSSD remained almost the same (P>0.05), while SDNN observably increased (P<0.05). It shows that the sum of HRV enlarged. (2) LF observably increased and HF observably dropped (P<0.05). It indicates that the function of sympathetic nerve boosted, while pneumogastric nerve wakened. (3) Correlation dimension and Lyapunov exponent declined obviously (P<0.05). It implies that the complexity and chaos degree falls, and it needs less past value or random effect to function. All the changing trend of these indexes shows that HRV can be a means for driving fatigue evaluating. HRV can not only estimate whether the driver is fatigue, but also judge the driving fatigue degree approximately.According to the conclusion, the estimate standard of vehicle driver's driving fatigue by HRV can be gained. This means use these three methods at the same time, make use of merits of linear and non-linear analysis. It makes the fatigue evaluation more exact and so it's a valuable technique. Study of HRV in the application of the monitoring of vehicle driver's fatigue has wide practice application and study foreground.