数据驱动跟驰模型综述

交通运输系统工程与信息 ›› 2021, Vol. 21 ›› Issue (5): 102-113.

数据驱动跟驰模型综述

贺正冰^*1，徐瑞康¹，谢东繁²，宗芳³，钟任新⁴

1. 北京工业大学，交通工程北京重点实验室，北京 100124；2. 北京交通大学，交通系统科学与工程研究院，北京 100044； 3. 吉林大学，交通学院，长春130022；4. 中山大学，智能工程学院，广州 510006

收稿日期:2021-04-08 修回日期:2021-04-16 接受日期:2021-04-22 出版日期:2021-10-25 发布日期:2021-10-21
作者简介:贺正冰(1982- )，男，辽宁锦州人，教授，博士。
基金资助:
国家重点研发计划

A Review of Data-driven Car-following Models

HE Zheng-bing^{* 1} , XU Rui-kang¹ , XIE Dong-fan² , ZONG Fang³ , ZHONG Ren-xin⁴

1. Beijing Key Laboratory of Traffic Engineering, Beijing University of Technology, Beijing 100124, China; 2. Institute of Transportation System Science and Engineering, Beijing Jiaotong University, Beijing 100044, China; 3. Transportation College, Jilin University, Changchun 130022, China; 4. School of Intelligent Systems Engineering, Sun Yat-Sen University, Guangzhou 510006, China

Received:2021-04-08 Revised:2021-04-16 Accepted:2021-04-22 Online:2021-10-25 Published:2021-10-21
Supported by:
National Key Research and Development Program of China

摘要/Abstract

摘要： 车辆跟驰模型是被交通科学与交通工程领域广泛认可的微观交通流模型，是交通流理论的基础。近年来，信息感知与获取、大数据、人工智能等技术快速发展，推动了数据驱动跟驰模型的快速发展。数据驱动跟驰模型，是以真实的车辆行驶数据为基础，利用数据科学与机器学习等理论和方法，通过样本数据的训练、学习、迭代、进化，挖掘车辆跟驰行为的内在规律。本文系统回顾了数据驱动跟驰模型在过去20余年的发展历程以及由神经网络和深度学习带动的两次研究热潮，归纳了基于传统机器学习理论的跟驰模型、基于深度学习的跟驰模型、模型与数据混合驱动的跟驰模型3类数据驱动跟驰模型，并分别介绍了其中的典型代表。分析数据源发现，尽管各种高精度轨迹数据不断涌现，目前研究仍多使用美国于2006年发布的Next Generation Simulation (NGSIM)高精度车辆轨迹数据，模型的可移植性和泛化能力值得思考与研究。提出关于模型输入、输出的3个问题：如何考虑更多驾驶行为变量，是否有必要考虑更多行为变量，现有输入、输出是否可替换。在模型测试与验证方面，发现并讨论了目前测试不充分、对比不完整、缺少统一测试集与测试标准等问题。最后，探讨了数据驱动跟驰模型原创性与成功的关键因素等问题。期望通过本文的梳理，帮助研究者更好地了解数据驱动跟驰模型的过去与现状，促进相关研究的快速发展。

关键词: 交通工程, 交通流理论, 深度学习, 机器学习, 大数据, 跟驰模型

Abstract: A car- following model is one of the microscopic traffic flow models that are widely focused on by transportation research and engineering. In recent years, the rapid technological advancement in information perception and acquisition, big data, and artificial intelligence, etc., has promoted the great development of data- driven carfollowing models. Based on data science and machine learning theory, data-driven car- following models obtain the inherent law of car- following behaviors through the training, learning, iteration and evolution of real-world vehicle motion data. This paper reviews the evolution of data-driven car-following models over the past 20 years and analyzes its two research waves driven by neural network and deep learning, respectively. Three typical types of data-driven carfollowing models and their representatives are reviewed, including traditional machine learning- based car- following models, deep learning-based car- following models, and model-data hybrid driven car- following models. Data source analysis indicates that, although a variety of high-fidelity trajectory datasets are constantly emerging, the Next Generation Simulation (NGSIM) datasets released by the United States in 2006 are still the most widely used, inparticular in recent years. Therefore, the transferability and generalization of the models are worth investigating. We also discuss from the following aspects: model input and output including how to involve more driving behavior variables, whether it is necessary to consider more behavioral variables, and whether the existing input and output can be replaced; Model testing and verification including insufficient testing, incomplete comparison, lack of unified test dataset and test standard. At last, the key factors regarding the originality and success of data- driven car- following models are discussed. It is expected that this review can help researchers better understand the past and present situations of data-driven car-following models and promote the progress of related research.

Key words: traffic engineering, traffic flow theory, deep learning, machine learning, big data, car-following model

中图分类号:

U491.2

贺正冰, 徐瑞康, 谢东繁, 宗芳, 钟任新. 数据驱动跟驰模型综述[J]. 交通运输系统工程与信息, 2021, 21(5): 102-113.

HE Zheng-bing , XU Rui-kang , XIE Dong-fan , ZONG Fang , ZHONG Ren-xin. A Review of Data-driven Car-following Models[J]. Journal of Transportation Systems Engineering and Information Technology, 2021, 21(5): 102-113.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.tseit.org.cn/CN/abstract/abstract30312.shtml

http://www.tseit.org.cn/CN/Y2021/V21/I5/102

参考文献

[1] BRACKSTONE M, MCDONAL M. Car-following: A historical review[J]. Transportation Research Part F: Traffic Psychology and Behaviour, 1999, 2(4): 181-196.

[2] CEDER A. The accuracy of traffic flow models: A review and preliminary investigation[J]. Traffic Engineering & Control, 1978, 11, 541-544.

[3] GIPPS P G. A behavioural car- following model for computer simulation[J]. Transportation Research Part B: Methodological, 1981, 15(2): 105-111.

[4] HANKEN A, ROCKWELL T H. A model of car following derived empirically by piece-wise regression analysis[C]// In Vehicular Traffic Science, International Symposium on the Theory of Traffic Flow, 1967.

[5] MICHACLS R M. Perceptual factors in car-following[J]. In Proceedings of the Symposium on Theory of Traffic Flow,OECD, 1963: 44-59.

[6] KIKUCHI S, CHAKROBORTY P. Car- following model based on fuzzy inference system[J]. Transportation Research Record, 1992, 1365: 82-91.

[7] HOOGENDOORN S P, BOVY P H L. State-of-the-art of vehicular traffic flow modelling[J]. Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, 2001, 215(4): 283- 303.

[8] OLSTAM J J, TAPANI A. Comparison of car- following models[R]. Stockholm: Swedish National Road andTransport Research Institute, 2004.

[9] PANWAI S, DIA H. Comparative evaluation of microscopic car-following behavior[J]. IEEE Transactions on Intelligent Transportation Systems, 2005, 6(3): 314- 325.

[10] TOLEDO T. Driving behaviour: models and challenges[J]. Transport Reviews, 2007, 27(1): 65-84.

[11] 王殿海, 金盛. 车辆跟驰行为建模的回顾与展望[J]. 中国公路学报, 2012, 25(1): 115-127. [WANG D H, JIN S. A review and prospect of vehicle following behavior modeling[J]. China Journal of Highway and Transport, 2012, 25(1): 115-127.]

[12] 杨龙海, 张春, 仇晓赟. 车辆跟驰模型研究进展[J]. 交通运输工程学报, 2019, 19(5): 125-138. [YANG L H, ZHANG C, QIU X Y. Research progress of vehicle following model[J]. Journal of Traffic and Transportation Engineering, 2019, 19(5): 125-138.]

[13] KEHTARNAVAZ N, GROSWOLD N, MILLER K, et al. A transportable neural- network approach to autonomous vehicle following[J]. IEEE Transactions on Vehicular Technology, 1998, 47(2): 694-702.

[14] 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(4): 92-94. [JIA H F, JUAN Z C, WANG X Y. The establishment of vehicle following model based on neural network[J]. Highway Traffic Science and Technology, 2001(4): 92-94.]

[15] MA X. A neural-fuzzy framework for modeling carfollowing behavior[C]. IEEE, Taipei，Taiwan, 2006.

[16] 徐学明, 荣建, 王丽. 混合神经网络跟驰模型的建立 [J]. 公路交通科技, 2007(3): 130-132. [XU X M, RONG J, WANG L. The establishment of the hybrid neural network following model[J]. Journal of Highway and Transportation Science and Technology, 2007(3): 130- 132.]

[17] PANWAI S, DIA H. Neural agent car-following models [J]. IEEE Transactions on Intelligent Transportation Systems, 2007, 8(1): 60-70.

[18] WEI D, LIU H. Analysis of asymmetric driving behavior using a self-learning approach[J]. Transportation Research Part B: Methodological, 2013, 47: 1-14.

[19] ZHANG J, LIAO Y, WANG S, et al. Study on driving decision-making mechanism of autonomous vehicle based on an optimized support vector machine regression [J]. Applied Sciences, 2018, 8(1): 13.

[20] HE Z, ZHENG L, GUAN W. A simple nonparametric carfollowing model driven by field data[J]. Transportation Research Part B: Methodological, 2015, 80: 185-201.

[21] ZHOU M, QU X, LI X. A recurrent neural network based microscopic car following model to predict traffic oscillation[J]. Transportation Research Part C: Emerging Technologies, 2017, 84: 245-264.

[22] WANG X, JIANG R, LI L, et al. Capturing car-following behaviors by deep learning[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 19(3): 910- 920.

[23] WANG X, JIANG R, LI L, et al. Long memory is important: A test study on deep-learning based carfollowing model[J]. Physica A: Statistical Mechanics and its Applications, 2019, 514: 786-795.

[24] HUANG X, SUN J, SUN J. A car- following model considering asymmetric driving behavior based on long short-term memory neural networks[J]. Transportation Research Part C: Emerging Technologies, 2018, 95: 346- 362.

[25] LIN Y, WANG P, ZHOU Y, et al. Platoon trajectories generation: A unidirectional interconnected LSTM-based car-following model[J]. IEEE Transactions on Intelligent Transportation Systems, 2020(99): 1-11.

[26] MA L, QU S. A sequence to sequence learning based carfollowing model for multi-step predictions considering reaction delay[J]. Transportation Research Part C: Emerging Technologies, 2020, 120: 102785.

[27] ZHU M, WANG X, WANG Y. Human-like autonomous car-following model with deep reinforcement learning[J]. Transportation Research Part C: Emerging Technologies, 2018, 97: 348-368.

[28] ZHOU Y, FU R, WANG C, et al. Modeling car-following behaviors and driving styles with generative adversarial imitation learning[J]. Sensors, 2020, 20(18): 5034.

[29] 丁点点, 孙磊, 陈松. 机器学习—动力学耦合车辆跟驰模型[J]. 交通运输系统工程与信息, 2017, 17(6): 33- 39. [DING D D, SUN L, CHEN S. Machine learningdynamic coupled vehicle following model[J]. Transportation Systems Engineering and Information Technology, 2017, 17(6): 33-39.

[30] 罗颖, 秦文虎. 基于IDM与RBFNN的组合型车辆低速跟驰模型[J]. 计算机应用研究, 2020, 37(8): 2354- 2357, 2380. [LUO Y, QIN W H. Low speed trailing model of combined vehicle based on IDM and RBFNN [J]. Computer Application Research, 2020, 37(8): 2354- 2357, 2380.]

[31] LI Y, LU X, REN C, et al. Fusion modeling method of carfollowing characteristics[J]. IEEE Access, 2019, 7: 162778-162785.

[32] 周志华, 机器学习[M]. 北京: 清华大学出版社, 2016 年. [ZHOU Z H. Machine learning[M]. Beijing: Tshinghua University Press, 2016.]

[33] U.S. Federal Highway Administration. The next generation simulation program (NGSIM) [EB/OL]. (2006- 12- 10) [2021- 04- 01]. http: // Ops. Fhwa. Dot. Gov/ Trafficanalysistools/Ngsim.Htm.

[34] LI L, JIANG R, HE Z, et al. Trajectory data-based traffic flow studies: A revisit[J]. Transportation Research Part C: Emerging Technologies, 2020, 114: 225-240.

[35] HE Z, ZHANG W, JIA N. Estimating carbon dioxide emissions of freeway traffic: A spatiotemporal cell-based model[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(5): 1976-1986.

[36] PUNZO V, MONTANINO M, CIUFFO B. Do we really need to calibrate all the parameters? Variance-based sensitivity analysis to simplify microscopic traffic flow models[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(1): 184-193.

[37] NEWELL G F. A simplified car-following theory: A lower order model[J]. Transportation Research Part B: Methodological, 2002, 36(3): 195-205.

[38] YU H, JIANG R, HE Z, et al. Automated vehicleinvolved traffic flow studies: A survey of assumptions, models, speculations, and perspectives[J]. Transportation Research Part C: Emerging Technologies, 2021, 127: 103101.