APP下载

基于大数据的矿用卡车驾驶风格识别算法研究

2021-08-17胡浈张瑞新刘鑫冯读康张秋涵

软件 2021年3期

胡浈 张瑞新 刘鑫 冯读康 张秋涵

摘 要:卡车司机的驾驶风格表征着其根据实时行车环境对矿车的运行控制所采取的一系列操作行为。对于矿用卡车安全性与经济性有较大影响。为提高驾驶风格聚类算法的适用性,并直观表达聚类效果,提出基于层次聚类的驾驶风格识别方法,结合实际采集的现场不同驾驶员的驾驶行为数据,进行驾驶风格识别。结果表明将驾驶风格分成3类的分类结果较为明显,且适合矿用卡车的驾驶风格识别。

关键词:卡车司机;驾驶风格;矿用卡车;层次聚类

中图分类号:TP393 文献标识码:A DOI:10.3969/j.issn.1003-6970.2021.03.005

本文著录格式:胡浈,张瑞新,刘鑫,等.基于大数据的矿用卡车驾驶风格识别算法研究[J].软件,2021,42(03):019-021+064

Research on Mining Truck Driving Style Recognition Algorithm Based on Big Data

HU Zhen, ZHANG Ruixin, LIU Xin, FENG Dukang, ZHANG Qiuhan

(School of Safety Engineering, North China Institute of Science and Technology, Beijing  065201)

【Abstract】:The driving style of a truck driver represents a series of operation behaviors taken by the driver to control the operation of the vehicle according to the real-time driving environment. This has a greater impact on the safety and economy of mining trucks. In order to improve the applicability of the driving style clustering algorithm and express the clustering effect intuitively, I propose a driving style recognition method based on hierarchical clustering, combining the actual collected driving behavior data of different drivers on-site to identify the driving style. The results show that the classification results of dividing the driving style into three categories are more obvious, and it is suitable for the driving style recognition of mining trucks.

【Key words】:truck driver;driving style;mining truck;hierarchical clustering

0 引言

驾驶风格用来表征驾驶员在实车运行环境下对车辆运行进行控制的操作行为特征,通过驾驶员操作习惯和汽车行驶数据的分析,动态识别出驾驶员的驾驶风格,对改善车辆燃油经济性有重要意义[1]。

对此,国内外学者运用不同技术手段结合不同卡车不同工况进行了一系列研究。吴振昕等[2]利用k-means聚类方法及D-S证据理论决策融合方法识别不同工况下驾驶风格。王超等[3]以驾驶员的视野特征和决策意愿表征驾驶风格应用Simulink/Carsim聯合仿真技术对驾驶员模型进行研究。胡杰等[4]对提出一种关联维数的驾驶风格指数,量化驾驶激进程度,从而精准识别驾驶风格。Kedar-Dongarkar等[5]提出一种基于车辆加速,制动,超速指数,油门指数的高效分类器,把驾驶风格分成激进、保守、适中3类。……

登录APP查看全文