摘要
通过智能移动终端内部的加速度传感器、陀螺仪传感器采集车辆运动状态数据,对采集的传感器数据进行移动平均滤波处理,提取传感器数据的时域特征.采用主成分分析对21维原始驾驶行为特征进行转换,生成五个表征明确驾驶行为的综合特征,结合k-means和模糊C均值(FCM)聚类算法对采集到的驾驶行为数据的综合特征进行聚类分析.聚类结果表明,车辆驾驶行为特征可有效聚为转弯行为、变速行为和变道行为,FCM聚类效果优于k-means,但是FCM的运行时间慢于k-means.
The acceleration sensor and gyroscope sensor inside the intelligent mobile terminal were used to collect the vehicle motion state data,and the collected sensor data was filtered by moving average to extract the time domain characteristics of the sensor data.The principal component analysis was adopted to transform the original driving behavior features of 21 dimensions to generate 5 comprehensive features.Combined with k-means and fuzzy C-means(FCM)clustering algorithm,the comprehensive characteristics of the collected driving behavior data were clustered and analyzed..The clustering results show that the vehicle driving behavior characteristics can be effectively clustered into turning behavior,speed changing behavior and lane changing behavior.FCM clustering effect is better than kmeans,but FCM running time is slower than k-means.
引文
[1]公安部交通管理局.《道路交通安全法》实施十三年驾驶人增两倍肇事率降八成[EB/OL].[2017-05-04].http:∥www.mps.gov.cn/n2254098/n4904352/c5692807/content.html.
[2]彭金栓,詹盛,徐磊,等.基于神经网络和贝叶斯滤波器的危险驾驶行为识别[J].武汉理工大学学报,2013,35(11):85-89.
[3]TAKENAKA K,BANDO T,NAGASAKA S,et al.Drive video summarization based on double articulation structure of driving behavior[C].In:ACM Multimedia,Seoul,2012.
[4]SATHYANARAYANA A,SADJADI S O,HANSEN J H L.Leveraging sensor information from portable devices towards automatic driving maneuver recognition[C].In:International IEEE Conference on Intelligent Transportation Systems, Yokohama,2012.
[5]JOHNSON D A,TRIVEDI M M.Driving style recognition using a smartphone as a sensor platform[C].In:International IEEE Conference on Intelligent Transportation Systems,Qingdao,2011.
[6]DIAZ A A,SERRADILLA G F,NARANJO J E,et al.Modeling the driving behavior of electric vehicles using smartphones and neural networks[J].IEEE Intelligent Transportation Systems Magazine,2014,6(6):44-53.
[7]EREN H,MAKINIST S,AKIN E,et al.Estimating driving behavior by a smartphone[C].In:Intelligent Vehicles Symposium,Changshu,2012.
[8] MILETTE G,STROUD A.Professional android sensor programming[M].北京:清华大学出版社,2013.
[9]宋雅清.基于Android传感器的驾驶事件识别关键技术研究[D].青岛:中国海洋大学,2014.
[10]肖献强,任春燕,王其东.机动车驾驶员驾驶行为操纵动作模式的研究[J].汽车工程,2015(9):1084-1088.
[11]郭孜政,陈崇双,王欣,等.基于FCM的驾驶行为险态辨识模型[J].吉林大学学报(工学版),2010,40(2):427-430.
[12]BENSAID A M,HALL L O,BEZDEK J C,et al.Validity-guided(re)clustering with applications to image segmentation[J].IEEE Transactions on Fuzzy Systems,2002,4(2):112-123.
[13]GIROLAMI M.Mercer kernel-based clustering in feature space[J].IEEE Transactions on Neural Networks,2002,13(3):780-790.
