中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2020, Vol. 39 ›› Issue (08): 1-5.DOI: 10.3969/j.issn.1674-0696.2020.08.01

• 交通+大数据人工智能 •    下一篇

遗传算法优化支持向量机的城市交通状态识别

李巧茹1,2,郝恩强1,陈亮1,2,范忠国1,杨文伟1   

  1. (1. 河北工业大学 土木与交通学院,天津 300401; 2. 河北工业大学 智慧基础设施研究院,天津 300401)
  • 收稿日期:2018-11-25 修回日期:2019-05-14 出版日期:2020-08-18 发布日期:2020-08-25
  • 作者简介:李巧茹(1972—),女,河北无极人,副教授,博士,主要从事道路交通管理与科学方面的研究。E-mail:qiaoruli129@126.com 通信作者:陈亮(1978—),男,天津人,副教授,主要从事道路交通管理与科学方面的研究。E-mail: karlchen@126.com
  • 基金资助:
    国家自然科学基金项目(51678212);河北省高等学校科学技术研究项目(QN2018231)

Urban Traffic State Recognition Based on Genetic Algorithm Optimized Support Vector Machine

LI Qiaoru1,2, HAO Enqiang1, CHEN Liang1,2, FAN Zhongguo1, YANG Wenwei1   

  1. (1. School of Civil Engineering and Transportation, Hebei University of Technology, Tianjin 300401, China; 2. Smart Infrastructure Research Institute, Hebei University of Technology, Tianjin 300401, China)
  • Received:2018-11-25 Revised:2019-05-14 Online:2020-08-18 Published:2020-08-25

摘要: 城市交通状态识别是智能交通控制、诱导和协同系统的基础。为提高支持向量机(support vector machine, SVM)在城市交通状态识别研究方面的泛化能力,将遗传算法(genetic algorithm, GA)与支持向量机相结合,利用遗传算法全局搜索优势对支持向量机的关键参数——惩罚系数C和核函数参数σ进行优化,建立基于遗传算法优化支持向量机(GA-SVM)的城市交通状态识别模型,并在MATLAB平台下进行实例验证。研究结果表明:相较于SVM模型,GA-SVM模型克服了依靠经验确定参数方法的缺点,识别精度提高3.75%,即模型可更好地识别城市交通状态。

关键词: 交通运输工程, 遗传算法, 支持向量机, 城市道路交通, 交通状态, 模式识别

Abstract: Urban traffic state recognition is the foundation of the intelligent traffic control, induction and coordination system. In order to improve the generalization ability of support vector machine in urban traffic state recognition, Genetic Algorithm (GA) and Support Vector Machine (SVM) were combined, and the key parameters of the support vector machine — the penalty coefficient C and the kernel function parameters σ were optimized by using the global search advantages of genetic algorithm. The urban traffic state recognition model based on genetic algorithm optimized support vector machine (GA-SVM) was established and verified by an example in MATLAB platform. The research results show that: compared with the SVM model, the GA-SVM model overcomes the shortcomings of determining the parameter method by means of experience, and the recognition accuracy is improved by 3.75%, that is to mean, the proposed model can better identify the urban traffic state.

Key words: traffic and transportation engineering, genetic algorithm, support vector machine, urban road traffic, traffic condition, pattern recognition

中图分类号: