摘要
本文通过对驾驶员的肌电信号与心电信号的研究,检测驾驶员驾车过程中的疲劳状态。对8名被试者进行2h的驾驶模拟实验。利用可穿戴式传感器采集被试者股二头肌部位的生理信号,采用快速独立成分分析和经验模态分解算法对测得的信号进行分离和去噪处理,得到肌电、心电信号,并找出能表征驾驶员疲劳的肌电和心电特性参数,运用统计分析SPSS软件进行Kolmogorov-Smirnov Z检验,最终选取肌电信号峰值因数和肌心电信号互相关峰值(P<0.001)作为组合特征,并采用马氏距离作为判别疲劳的准则。结果表明,该方法在对驾驶员正常状态与疲劳状态的区分上有良好的识别效果。
The fatigue states of drivers during driving are detected based on the study on the EMG and ECG signals of drivers in this paper. Firstly eight testees take part in an experiment by driving on driving simulator for two hours, putting on wearable sensor to collect their physiological signals of biceps femoris. Then separation and denoising treatments are conducted on the signals collected with FastlCA and EMD algorithm to obtain the EMG and ECG signals, and the electrocardiographic and electromyographic characteristic parameters representing the extent of driver＇s fatigue are found and subjected to Kolmogorov-Smirnov Z test by statistical software SPSS. Finally the peak value factor of EMG signal and the peak value of correlation between electrocardiographic and electromyographic sig- nals are selected as combined features with Mahalanobis distance as criteria for fatigue judgment. The results show that the method proposed has good identification effect in distinguishing the normal and fatigue states of drivers.
出处
《汽车工程》
EI
CSCD
北大核心
2013年第12期1143-1148,共6页
Automotive Engineering
基金
国家自然科学基金(61071057)
中央高校基本科研业务费项目(N100603003)资助
