基于SVM算法对车辆加速踏板舒适性的研究与实现  

Research and implementation of vehicle accelerator pedal comfort based on SVM algorithm

作　　者：孔尧[1] 曹曦文 张育华 曾洁[3] 宋雪桦[1] KONG Yao;CAO Xi-wen;ZHANG Yu-hua;ZENG Jie;SONG Xue-hua(School of Computer Science and Communication Engineering,Jiangsu University,Zhenjiang 212000,Jiangsu Province,China;Hengchi Technology Co.,L td.,Zhenjiang 212200,Jiangsu Province,China;School of Electrical Information,Dalian Jiaotong University,Dalian 116000,Liaoning Province,China)

机构地区：[1]江苏大学计算机科学与通信工程学院,江苏镇江212000 [2]江苏恒驰科技有限公司,江苏镇江212200 [3]大连交通大学电气信息学院,辽宁大连116000

出　　处：《信息技术》2018年第6期22-25,共4页Information Technology

基　　金：江苏省科技支撑计划项目(BE2015132);2016年辽宁省自然基金(201602116)

摘　　要：为了改善车辆加速性能从而提高司机驾驶舒适度,文中提出利用分类算法对车辆加速性能进行研究。首先将采集到的踏板开度数据预处理,其次进行特征提取,通过支持向量机分类算法有效提高加速踏板开度分类精度,并运用Hadoop分布式处理方式来提高大量分类数据处理的效率。测试结果表明,文中提出的方法提高了分类准确性提高到70%以上以及hadoop集群分类效率提高显著。In order to improve vehicle acceleration performance and improve driver ＇s driving comfort,this paper proposes a classification algorithm to study vehicle acceleration performance. First of all the collected pedal in data preprocessing,feature extraction and classification algorithm of support vector machine,it improves the accelerator opening classification accuracy,and the use of Hadoop distributed processing mode improves the efficiency of a large number of classified data processing. The test results show that the proposed method improves the classification accuracy to more than 70%, and the classification efficiency of Hadoop cluster is improved significantly.

关 键 词：加速踏板开度 SVM HADOOP 分布式 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]