摘要
为了对驾驶员的驾驶状态进行有效地监测,基于驾驶员握力、方向盘转角和驾驶员头部位置3种驾驶信息,通过分析驾驶过程中的内在规律,提出了一种滑动时间窗的提取方法,使用该方法提取出了握力幅值疲劳度、转角标准差疲劳度、转角频数疲劳度和头位偏离疲劳度4种特征参数,并结合神经网络技术,建立了驾驶状态实时监测系统。研究表明:所设计的驾驶状态实时监测方法的识别正确率可达90.4%,且监测系统实时性强、监测效果理想、获取信息直观、警示方式多功能化,为驾驶状态的特征提取和实时监测系统的研究提供了有益的参考。
This paper presents a new method to effectively monitor the driving state.Based on the grip force of driver,the steering wheel angle and the head position,through the analysis of the inherent law in the process of driving,an extracting method of sliding time window was proposed,by which four characteristic parameters：the fatigue of the grip force amplitude,the fatigue of the angle standard deviation,the fatigue of the angle frequency and the fatigue of the head position deviation,were extracted.Combined with artificial neural network,the real-time monitoring system was constructed.The results show that the accuracy rate of the presented monitoring method can reach 90.4 percent,and this monitoring system has the features of good real-time performance,good monitoring effect,obtain information directly and multi-functional warning,which has great reference significance for the fatigue characteristic extraction and the real-time monitoring system.
出处
《电子测量与仪器学报》
CSCD
2014年第9期965-973,共9页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61072032)项目
关键词
特征参数
滑动时间窗
实时监测
疲劳报警
characteristic parameters
sliding time window
real-time monitoring
fatigue warning
