摘要
为了提高运输安全管理水平和运输效率,对运输车辆安全性进行客观评价,提出了一种结合多算法的行车安全评价模型。首先,根据交通运输部公路科学研究院所给车联网数据,设计并定义了驾驶人行为的特征指标和评价指标;其次,采用回归分析方法对驾驶人的驾驶风格进行分析;接着,分别采用K-means聚类和DBSCAN聚类算法对驾驶人的不良行为进行分析;经过对比,最终选用K-means聚类算法和因子分析的结果对驾驶行为进行评价,共分为6类。该模型将机器学习中的数据挖掘和数据分析算法与道路运输行业相结合,为道路运输安全管理的研究提供了一个量化分析的工具。
In order to improve the transportation safety management level and transportation efficiency, make an objective evaluation of the vehicles safety, a driving safety evaluation model combined with the multiple algorithms is proposed. First of all, according to the data of the internet-connected vehicles provided by the Highway Science Research Institute of the Ministry of Transport, the characteristics and evaluation indicators of the driver behaviors are designed and defined. Secondly, the regression analysis method is used to analyze the driving style. Then, the K-means clustering and DBSCAN clustering algorithms are used to analyze the driver′s unsafe behaviors. After the comparison, the results of K-means clustering algorithm and factor analysis are used to evaluate the driving behaviors of 6 categories. Combining the data mining and the data analysis algorithms in machine learning with the road transport industry, a quantitative analysis tool for the study of road transport safety management is proposed.
作者
李卓轩
林凯迪
郭建华
曹进德
LI Zhuoxuan;LIN Kaidi;GUO Jianhua;CAO Jinde(School of Mathematics,Southeast University,Nanjing 210096,China;School of Science,Shenyang University of Technology,Shenyang 110870,China;ITS Research Center,Southeast University,Nanjing 210096,China)
出处
《南通大学学报(自然科学版)》
CAS
2020年第1期26-32,47,共8页
Journal of Nantong University(Natural Science Edition)
基金
国家自然科学基金项目(61573096,61833005)
江苏省网络群体智能重点实验室项目(BM2017002)。