摘要
交通违法是导致交通事故的重要因素,从驾驶者发生交通事故类型及其历史交通违法类型的角度研究危险驾驶行为。利用归一化和阈值法,选取合适的交通违法类型;通过χ~2检验交通违法和交通事故两者关联性;再采用对应分析的方法,研究不同交通事故类型对应的主次要相关违法类型,并建立影响程度模型;同时,基于两者的交互影响,对部分交通事故类型和违法类型分类、合并。结果表明:交通事故类型和交通违法类型间有显著的相关性和对应关系,基于这种特性可预防交通事故,预判事故类型。
Illegal driving usually results in traffic accidents.This paper was aimed at exploring the internal relation between illegal driving and traffic accident by analyzing the drivers' historical behavior.13 types of illegal driving with relatively high frequencies and abnormal expected values were identified from 184 types of illegal driving by using normalization and threshold method.The correlations between illegal driving and traffic accidents were demonstrated by chi-square test.An influence degree model was built for correspondence between types of traffic accidents and types of illegal driving by the corresponding analysis method.The results show that there is a significant correlation between types of traffic accidents and types of illegal driving,and that for a given driver,a probable traffic accident type can be predicted from his or her driving experience.
作者
刘林
傅惠
吕伟韬
胡刚
王晨
张喆
LIU Lin;FU Hui;LYU Weitao;HU Gang;WANG Chen;ZHANG Zhe(School of Electro-Mechanical Engineering, Guangdong University of Technology, Guangzhou Guangdong 510006, China;Jiangsu Intelligent Transportation Company Limited, Nanjing Jiangsu 211111, China;Intelligent Transportation System (ITS) Research Center at Southeast University, Nanjing Jiangsu 210096, China;Public Security Bureau Traffic Police of Kunshan North Squadron, Suzhou Jiangsu 215000, China)
出处
《中国安全科学学报》
CAS
CSCD
北大核心
2017年第11期79-84,共6页
China Safety Science Journal
基金
国家自然科学基金资助(61573110)
广州市科技计划项目(201504291250033)