基于大数据的智能辅助驾驶道路适用度评估  

Road Suitability Evaluation of Intelligent Assistant Driving Based on Machine Learning

作　　者：谢燕芳 张亮 程登 郑陈亮 卓丽 XIE Yanfang;ZHANG Liang;CHENG Deng;ZHENG Chenliang;ZHUO Li(SAIC GM Wuling Automobile Co.,Ltd.,Guangxi Laboratory of New Energy Automobile,Guangxi Key Laboratory of Automobile Four New Features,Liuzhou 545007,China)

机构地区：[1]上汽通用五菱汽车股份有限公司,广西新能源汽车实验室,广西汽车新四化重点实验室,广西柳州545007

出　　处：《汽车电器》2023年第9期16-18,共3页Auto Electric Parts

摘　　要：随着汽车市场朝网联化发展,主机厂可通过网联终端采集到智能辅助驾驶的行车数据,主机厂通过车联网大数据离线计算了解用户使用智能辅助驾驶功能的情况,分析智能辅助驾驶在各类路况中的适用性。汽车主机厂在研发阶段做了各路况的实车测试,但未有一套基于用户实际使用情况的评估方法。为了了解智能辅助驾驶功能在各类路况中的适用度,需要一套基于用户智能辅助驾驶数据的评估方法。本文基于车联网采集的智能辅助驾驶功能的数据,使用机器学习的方法对汽车智能辅助驾驶的道路适用度进行评估。With the development of the automobile market towards networking,the OEMs can collect the driving data of intelligent auxiliary driving through the networking terminals.The OEMs can understand the user's use of intelligent auxiliary driving functions through the offline calculation of the big data of the Internet of Vehicles,and analyze the applicability of intelligent auxiliary driving in various road conditions.In the R&D stage,the automobile OEMs have conducted real vehicle tests on various road conditions,but there is no evaluation method based on the actual use of users.In order to understand the applicability of intelligent driving assistance in various road conditions,an evaluation method based on user's intelligent driving assistance data is required.Based on the data of intelligent assistant driving function collected by the Internet of Vehicles,this paper uses the method of machine learning to evaluate the road suitability of intelligent assistant driving.

关 键 词：智能辅助行车 道路适用度 评估算法 路向聚类 数据网格化 

分 类 号：U463.6[机械工程—车辆工程]