Research on Driving Fatigue Level Using ECG Signal from Smart Bracelet

摘要

Fatigue level was studied using ECG signals collected from smart bracelet, and the data collected from the smart bracelet could reduce the interference to the driver during the experiment, increase the accuracy of the data. In order to accurately fatigue level, the algorithm of driving fatigue level was proposed based on multi-index fusion theory. Aiming at the missing status of the comprehensive indicator T value from principal component analysis, BP neural network was used for data compensation. The threshold of fatigue level was determined on the basis of the peak trough of the comprehensive index T value. Based on experimental analysis, it was found that the algorithm can effectively identify the wide awake and severe fatigue states when the data was missing.