基于BP神经网络的驾驶精神疲劳识别方法

Recognition method of driving mental fatigue based on BP neural network

摘要:

为了对驾驶精神疲劳予以有效识别,基于行为绩效结合心电信号指标构建了一种驾驶精神疲劳识别方法．以驾驶行为绩效为客观测评指标,给出了驾驶精神疲劳状态的分级划分方法．在此基础上,以心率变异性的6项指标作为疲劳识别特征因子,采用BP神经网络模型,建立了驾驶精神疲劳状态分类器．最后结合实例,依据驾驶行为绩效,将疲劳状态划分为2级,采用10名驾驶员连续4 h的驾驶行为绩效(反应时)、心电数据,对模型、方法予以测算．结果表明,10名驾驶员平均正确识别率在71％～80％之间,且其平均正确识别率为73％．BP神经网络模型与心率变异性指标相结合可有效的识别疲劳．

Abstract:

To recognize driving mental fatigue efficiently, this study constructs a recognition method based on ECG. The method proposes hierarchy partition of state of driving mental fatigue by using driving behavior performance as objective evaluation indexes. Meanwhile, taking 6 indexes of HRV as fatigue recognition characterization factors and BP artificial neural network model, this paper establishes the recognition model for state of driving mental fatigue. Finally, according to examples, the mental fatigue is divided into two classifications. Collecting 4 hours continual driving behavior performance and ECG data from 10 drivers to test the model, the result shows that the average recognition accuracy rate is between 71％ and 80％, and the average accuracy rate is 73％. The combination of BP neural network model and HRV indexes could recognize fatigue effectively.