A Car-Following Model Based on Support Vector Machine

doi:10.3969/j.issn.1674-0696.2015.06.24

Journal of Chongqing Jiaotong University(Natural Science) ›› 2015, Vol. 34 ›› Issue (6): 128-132.DOI: 10.3969/j.issn.1674-0696.2015.06.24

• Transportation Engineering • Previous Articles Next Articles

A Car-Following Model Based on Support Vector Machine

Qiu Xiaoping^1,2,3, Liu Yalong¹

1. School of Transportation & Logistics， Southwest Jiaotong University， Chengdu 610031， Sichuan， China; 2. Comprehensive Intelligent Transportation National & Local Joint Engineering Laboratory， Chengdu 610031， Sichuan， China; 3. Comprehensive Transportation Key Laboratory of Sichuan Province， Chengdu 610031， Sichuan， China

Received:2015-04-20 Revised:2015-07-08 Online:2015-12-30 Published:2015-12-29

基于支持向量机的车辆跟驰模型

邱小平^1,2,3，刘亚龙¹

1. 西南交通大学交通运输与物流学院，四川成都 610031；2. 综合交通运输智能化国家地方联合工程实验室，四川成都 610031；3. 综合运输四川省重点实验室，四川成都 610031

作者简介:邱小平(1976—)，男,四川南充人，教授，工学博士，主要从事交通运输规划与管理方面的研究。E-mail: qxp@home.swjtu.edu.cn。
基金资助:
国家自然科学基金项目(51278429, 51408509)；四川省科技厅项目(2013GZX0167, 2014ZR0091)；中央高校基本业务经费项目(SWJTU11CX080)；成都市科技局项目(2014-RK00-00056-ZF)

Abstract

Abstract: A car following model based on support vector machine algorithm was established to simulate the car-following behavior:acceleration, deceleration, no action. The SVM car-following model was trained and tested using Next Generation Simulation (NGSIM) data, and then the test results were compared with those obtained from Gipps car-following model. The results indicate that comparing with Gipps model, the accuracy of error indicators of the SVM car-following model is greatly improved; moreover, SVM car-following model can explore the potential correlationship between variables impacting car following behavior, which makes up for the deficiencies of the traditional car-following model.

Key words: traffic and transportation engineering, car-following model, machine learning, support vector machine, regression forecast

摘要： 基于支持向量机算法建立车辆跟驰模型，模拟单车道车辆跟驰行为——加速、减速、无动作；利用NGSIM数据对模型进行训练和测试，并与Gipps车辆跟驰模型的测试结果进行对比。结果表明：所建模型各项误差指标的精度均有较大提升，能够挖掘出影响跟驰行为的变量之间的潜在关系，弥补了传统车辆跟驰模型的不足。

关键词: 交通运输工程, 车辆跟驰模型, 机器学习, 支持向量机, 回归预测

CLC Number:

U491.2

Qiu Xiaoping, Liu Yalong. A Car-Following Model Based on Support Vector Machine[J]. Journal of Chongqing Jiaotong University(Natural Science), 2015, 34(6): 128-132.

邱小平，刘亚龙. 基于支持向量机的车辆跟驰模型[J]. 重庆交通大学学报（自然科学版）, 2015, 34(6): 128-132.

References

［1］ Gazis D C,Herman R,Rothery R W.Nonlinear following the leader models of traffic flow ［J］.Operations Research,1961,9(4):545-567.
［2］ Kometani E,Sasaki T.Dynamic behaviour of traffic with a nonlinear spacing-speed relationship ［C］// Proceedings of the Symposium on Theory of Traffic Flow.New York:Elsevier,1959.
［3］ Gipps G.A behavioural car-following model for computer simulation ［J］.Transportation Research Part B:Methodological,1981,15(2):105-111.
［4］ Wiedemann R.Simulation of Road Traffic in Traffic Flow ［R］.Karlsruhe:University of Karlsruhe (TH),1974.
［5］ Kikuchi C,Chakroborty P.Car following model based on a fuzzy inference system ［J］.Transportation Research Record,1992,1365:82-91.
［6］贾洪飞,隽志才,王晓原.基于神经网络的车辆跟驰模型的建立［J］.公路交通科技,2001,18(4):92-94.
Jia Hongfei,Juan Zhicai,Wang Xiaoyuan.Development of a car-following model based on artificial neural networks ［J］.Journal of Highway and Transportation Research and Development,2001,18(4):92-94.
［7］ Simonelli F,Bifulco G N,De Martinis V,et al.Human-like adaptive cruise control systems through a learning machine approach ［J］. Applications of Soft Computing,2009,52:240-249.
［8］ Cortes C,Vapnik V.Support-vector networks ［J］.Machine Learning,1995,20(3):273-297.
［9］ Drucker H,Burges C J C,Kaufman L,et al.Support vector regression machines ［J］. Advances in Neural Information Processing Systems ,1996,28(7):779-784.
［10］杨达,蒲云,杨飞,等.基于最优间距的车辆跟驰模型及其特性［J］.西南交通大学学报,2012,47(5):888-894.
Yang Da,Pu Yun,Yang Fei,et al.Car-following model based on optimal distance and its characteristics analysis ［J］.Journal of Southwest Jiaotong University,2012,47(5):888-894.
［11］Punzo V,Simonelli F.Analysis and comparison of microscopic traffic flow models with real traffic microscopic data ［J］.Transportation Research Record,2005,1934(1):53-63.
［12］Elmar Brockfeld,Reinhart D Kühne,Peter Wagner.Calibration and validation of microscopic models of traffic flow ［J］.Transportation Research Record,2005,1934(1):179-187.

[1]	LI Jing1,2, ZHONG Peng1, LYU Jing1,2. Interaction between Land Port and Urban Economy Based on System Dynamics [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 1-7.
[2]	LI Jun1, ZHU Yunsheng2. Application of Comprehensive Evaluation of Triangular Fuzzy Analytic Hierarchy Process in Ship Safety Assessment of the Yellow River Reservoir Area [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 16-22.
[3]	DENG Ping1,2, SONG Lian1, HUANG Chengfeng1,2. Estimation of Congestion Costs for Lockage Ship in Three Gorges Dam Area and Its Impact on the Total Value of Goods in the Dam Area [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 23-30.
[4]	CHENG Qian1, ZHU Xiaoning2, LU Wansheng3. Travel Mode Choice Behavior Model of Intercity Passengers with Medium and Long Haul — A Case Study of High-Speed Railway and Air Transport [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 39-45.
[5]	LIAN Qicai1, LI Han1, SHI Xiaolin1, YAN Zhangcun2. Autonomous Driving Choice Behavior Based on Panel Data Mixed Logit Model [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 46-52.
[6]	QI Chunhua, WANG Xiaonan, ZHU Shoulin, LI Hangtian. Traffic Engineering Facility Information Threshold Based on Drivers Visual Area of Interest [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 53-60.
[7]	LI Shengchao, WU Yue, BAI Xuemeng, WANG Yue, ZHANG Hai. Violation Alarm of Dangerous Goods Vehicle Based on TOPSIS and Gray Correlation [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 61-66.
[8]	YANG Qingfang1,2,3, WANG Liqiang2, ZHENG Lili1,2,3, MENG Fanyun2. Traffic Emission and Fuel Consumption Model of Roundabout [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(07): 67-73.
[9]	BAI Han, ZHANG Kangyu, WANG Xiuguang, YAN Xiang, FENG Qilong. Traffic Capacity of Urban Intersection under the Conditions of Mixed Release of Traffic Flow [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(06): 12-20.
[10]	ZHANG Chongyang, HUANG Shuping. Seismic Reliability of Road Network and Route Selection of Post-Earthquake Transportation [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(06): 43-51.
[11]	GUO Wenwen1, JI Mingjun2, ZHU Huiling3, ZHOU Wenjie2. Optimization Model of Yard Crane Scheduling in a Container Terminal [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(06): 66-72.
[12]	LI Guangru, ZHANG Xin, ZHU Qinghui. Prediction Method of Port Throughput with Improved Elman Neural Network Based on Adaboost [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(05): 1-5.
[13]	SUN Yuanguang1, JIAN Min2, JIN Hua2, CHEN Shaokuan2. Calculation and Simulation of Delay during Train Failure Rescue of Urban Rail Transit [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(05): 6-12.
[14]	WANG Honggang1,2, BAI Peng1, ZOU Qingru2, ZHAO Ling1. Optimization Model of Track Application for Metro Depots [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(05): 13-18.
[15]	SUN Yanhao1,2, ZHANG Qi1,2, REN Yumou1,2,3, SUN Pengju4. Evaluation of Emergency Disposal Ability of Train Dispatcher of High-Speed Railway [J]. Journal of Chongqing Jiaotong University(Natural Science), 2021, 40(05): 19-25.

A Car-Following Model Based on Support Vector Machine

基于支持向量机的车辆跟驰模型

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics