基于智能手表的疲劳驾驶监测方法研究  

Research on Driver Fatigue Monitoring Method Based on Smart Watch

作　　者：杨萍茹 黄勇[2] 廖龙涛 孙棣华[2] 陈希 王正江 YANG Pingru;HUANG Yong;LIAO Longtao;SUN Dihua;CHEN Xi;WANG Zhengjiang(Chongqing City Comprehensive Transportation Hub Development and Investment Co.,Ltd.,Chongqing 401121,China;School of Automation,Chongqing University,Chongqing 400044,China;Chongqing Public Transport Holding(Group)Co.,Ltd.,Chongqing 401121,China)

机构地区：[1]重庆城市综合交通枢纽开发投资有限公司,重庆401121 [2]重庆大学自动化学院,重庆400044 [3]重庆市公共交通控股(集团)有限公司,重庆401121

出　　处：《重庆理工大学学报（自然科学）》2020年第12期170-176,共7页Journal of Chongqing University of Technology：Natural Science

基　　金：重庆市科技计划项目基础科学与前沿技术研究专项(cstc2017jcyjBX0001)。

摘　　要：为了提升疲劳驾驶监测的实用性,采用日益普及的智能手表采集驾驶员转向行为数据信息,从中提取了多个有效疲劳特征指标,建立了基于随机森林的疲劳驾驶监测模型。通过在模拟驾驶平台上进行驾驶实验,使用智能手表采集驾驶员的转向行为数据,并以10 s的时间窗对数据进行特征提取,完成疲劳驾驶监测模型的训练与验证。测试结果表明:所提出监测方法的综合疲劳检测准确率达到85.27%,能够有效监测疲劳驾驶行为。In order to improve the practicability of fatigue driving monitoring,this paper uses the increasingly popular smart watch to collect driver steering behavior information,and extracts several effective fatigue characteristic indicators,and establishes a fatigue driving monitoring model based on random forest.Through driving experiment on the simulated driving platform,the data of driver’s turning behaviors are collected by smart watch,and the data are extracted by 10 s time window to complete the training and validation of fatigue driving monitoring model.The test results show that the comprehensive fatigue detection accuracy of this paper can reach 85.27%,which can effectively monitor the fatigue driving behavior.

关 键 词：疲劳驾驶 转向行为 智能手表 随机森林 

分 类 号：U471[机械工程—车辆工程] TP18[交通运输工程—载运工具运用工程]