基于多源数据融合的城市道路旅行时间预测
|
摘要
城市道路旅行时间预测是城市道路交通管理的重要支撑。研究了利用多源数据预测城市信号控制主干道旅行时间的方法。以2015年8月某路段的视频检测器及微波检测器的2种数据为基础,采用回归拟合的方法探究信号控制主干道路段旅行时间与断面流量之间的关系,2段式的线性拟合结果可以较好地拟合信号控制主干道路段旅行时间与断面流量的关系。以BP神经网络模型为基础,从输入层入手,采用直接输入2类数据、应用拟合关系输入拟合数据等方法,综合考虑2类数据之间的相关性,建立了融合2类检测数据进行旅行时间预测的多个模型,对7种不同输入的神经网络预测模型进行了测试、对比和分析。研究结果表明,相比于时间序列、支持向量机、k近邻和历史平均方法而言,应用拟合关系的2类数据融合的BP-2神经网络模型具有更高的预测精度,MAPE为13.04%,表明BP2神经网络模型能够实现较好的旅行时间预测效果。
Travel time prediction of urban road network is an important basis for traffic management. A method for travel time prediction is designed based on signalized arterial sections by integrating multi-source data. The license plate recognition data and microwave radar detection data of a signalized arterial section during August, 2015 is used to investigates relationships between travel time and traffic volume by conducting regression models. Results show that a two-stage linear regression model performs the best. Furthermore, on the basis of the Back Propagation(BP) neural network model, seven neural network models with different specifications are conducted, compared and analyzed. Specifically, the neural network models are established with different input layers that consider the relationships between different variables. Results show that the BP-2 neural network model, which considers the relationship between the two types of data, has higher prediction accuracy(MAPE=13.04%) than which of time series, support vector machine, k-nearest neighboring, and historical average models. It is proven that BP-2 neural network model can effectively predict travel time.
Travel time prediction of urban road network is an important basis for traffic management. A method for travel time prediction is designed based on signalized arterial sections by integrating multi-source data. The license plate recognition data and microwave radar detection data of a signalized arterial section during August, 2015 is used to investigates relationships between travel time and traffic volume by conducting regression models. Results show that a two-stage linear regression model performs the best. Furthermore, on the basis of the Back Propagation(BP) neural network model, seven neural network models with different specifications are conducted, compared and analyzed. Specifically, the neural network models are established with different input layers that consider the relationships between different variables. Results show that the BP-2 neural network model, which considers the relationship between the two types of data, has higher prediction accuracy(MAPE=13.04%) than which of time series, support vector machine, k-nearest neighboring, and historical average models. It is proven that BP-2 neural network model can effectively predict travel time.
引文
[1] DONG C,XIONG Z,SHAO C,et al.A spatial-temporal-based state space approach for freeway network traffic flow modelling and prediction[J].Transportmetrica A:Transport Science,2015,11(7):547-560.
[2] VLAHOGIANNI E I,KARLAFTIS M G,GOLIAS J C.Short-term traffic forecasting:Where we are and where we′re going[J].Transportation Research Part C:Emerging Technologies,2014(43):3-19.
[3] 冯微,陈红,张兆津,等.基于GBRBM-DBN模型的短时交通流预测方法[J].交通信息与安全,2018,36(5):99-108.FENG Wei,CHEN Hong,ZHANG Zhaojin,et al.A forecast of short-term traffic flow based on GBRBM-DBN model[J].Journal of Transport Information and Safety,2018,36(5):99-108.(in Chinese)
[4] 梅朵,郑黎黎,冷强奎,等.基于时空GPSO-SVE的短时交通流预测[J].交通信息与安全,2017,35(2):68-74,120.MEI Duo,ZHENG Lili,LENG Qiangkui,et al.A prediction model for short-term traffic flow based on space-time GPSO-SVM[J].Journal of Transport Information and Safety,2017,35(2):68-74,120.(in Chinese)
[5] WESTERMAN M,LITJENS R,LINNARTZ J P.Integration of probe vehicle and induction loop data:estimation of travel times and automatic incident detection[J].1996,933(4):1-10.
[6] RAN B,ROUPHAIL N M,TARKO A,et al.Toward a class of link travel time functions for dynamic assignment models on signalized networks[J].Transportation Research Part B:Methodological,1997,31(4):277-290.
[7] 崔梦莹,刘锴,候文宇,等.基于自适应加权平均融合的路段行程时间估计[J].交通标准化,2014,42(15):38-42.CUI Mengying,LIU Kai,HOU Wenyu,et al.Link travel time estimation based on adaptive weighted average fusion[J].Transportation Standardization,2014,42(15):38-42.(in Chinese)
[8] KOLAROV A,ATAI A,HUI J.Application of Kalman filter in high-speed networks[C].IEEE Global Telecommunications Conference,San Francisco,USA:The Global Bridge,IEEE,1994.
[9] 刘红红,杨兆升.基于数据融合技术的路段出行时间预测方法[J].交通运输工程学报,2008(6):88-92.LIU Honghong,YANG Zhaosheng.Estimating methods of link travel times based on data fusion technology[J].Journal of Traffic and Transportation Engineering,2008(6):88-92.(in Chinese)
[10] 董春娇,邵春福,周雪梅,等.基于交通流参数相关的阻塞流短时预测卡尔曼滤波算法[J].东南大学学报(自然科学版),2014,44(2):413-419.DONG Chunjiao,SHAO Chunfu,ZHOU Xuemei,et al.Kalman filter algorithm for short-term jam traffic prediction based on traffic parameter correlation[J].Journal of Southeast University (Natural Science Edition).2014,44(2):413-419.(in Chinese)
[11] CHOI K,CHUNG Y.A data fusion algorithm for estimating link travel time[J].Journal of Intelligent Transportation Systems,2002,7(3):235-260.
[12] ZHENG W,LEE D H,ZHENG W,et al.Short-term freeway traffic flow prediction:Bayesian combined neural network approach[J].Journal of Transportation Engineering,2006,132(2):114-121.
[13] 王建,邓卫,赵金宝.基于改进型贝叶斯组合模型的短时交通流量预测[J].东南大学学报(自然科学版),2012,42(1):162-167.WANG Jian,DENG Wei,ZHAO Jinbao.Short-term freeway traffic flow prediction based on improved Bayesian combined model[J].Journal of Southeast University (Natural Science Edition),2012,42(1):162-167.(in Chinese)
[14] PARK D,RILETT L R,HAN G.Spectral basis neural networks for real-time travel time forecasting[J].Journal of Transportation Engineering,1999,125(125):515-523.
[15] 邱奉翠.基于城市快速路和主干道多源交通检测信息的数据融合技术研究[D].北京:北京交通大学,2012.QIU Fengcui.Study on data fusion technology of multi-source traffic data of urban expressway and arterial road[D].Beijing:Beijing Jiaotong University,2012.(in Chinese)
[16] 姚恩建,张永生.城市快速路区间旅行时间短期预测算法[J].武汉理工大学学报(交通科学与工程版),2013,37(6):1133-1137.YAO Enjian,ZHANG Yongsheng.A new algorithm of short-term travel time prediction for urban expressway[J].Journal of Wuhan University of Technology (Transportation Science & Engineering),2013,37(6):1133-1137.(in Chinese)
[17] 孙祥龙,陆建.公交站点间行程时间估计[J].武汉理工大学学报(交通科学与工程版),2013,37(1):67-70.SUN Xianglong,LU Jian.Travel time estimation between bus stop[J].Journal of Wuhan University of Technology (Transportation Science & Engineering),2013,37(1):67-70.(in Chinese)
[18] 黄靖翔.基于多源数据融合的高速公路路段行程时间估计[D].广州:华南理工大学,2018.HUANG Jingxiang.Link travel time estimation of expressway based on multi-source data fusion[D].Guangzhou:South China University of Technology,2018.(in Chinese)
[2] VLAHOGIANNI E I,KARLAFTIS M G,GOLIAS J C.Short-term traffic forecasting:Where we are and where we′re going[J].Transportation Research Part C:Emerging Technologies,2014(43):3-19.
[3] 冯微,陈红,张兆津,等.基于GBRBM-DBN模型的短时交通流预测方法[J].交通信息与安全,2018,36(5):99-108.FENG Wei,CHEN Hong,ZHANG Zhaojin,et al.A forecast of short-term traffic flow based on GBRBM-DBN model[J].Journal of Transport Information and Safety,2018,36(5):99-108.(in Chinese)
[4] 梅朵,郑黎黎,冷强奎,等.基于时空GPSO-SVE的短时交通流预测[J].交通信息与安全,2017,35(2):68-74,120.MEI Duo,ZHENG Lili,LENG Qiangkui,et al.A prediction model for short-term traffic flow based on space-time GPSO-SVM[J].Journal of Transport Information and Safety,2017,35(2):68-74,120.(in Chinese)
[5] WESTERMAN M,LITJENS R,LINNARTZ J P.Integration of probe vehicle and induction loop data:estimation of travel times and automatic incident detection[J].1996,933(4):1-10.
[6] RAN B,ROUPHAIL N M,TARKO A,et al.Toward a class of link travel time functions for dynamic assignment models on signalized networks[J].Transportation Research Part B:Methodological,1997,31(4):277-290.
[7] 崔梦莹,刘锴,候文宇,等.基于自适应加权平均融合的路段行程时间估计[J].交通标准化,2014,42(15):38-42.CUI Mengying,LIU Kai,HOU Wenyu,et al.Link travel time estimation based on adaptive weighted average fusion[J].Transportation Standardization,2014,42(15):38-42.(in Chinese)
[8] KOLAROV A,ATAI A,HUI J.Application of Kalman filter in high-speed networks[C].IEEE Global Telecommunications Conference,San Francisco,USA:The Global Bridge,IEEE,1994.
[9] 刘红红,杨兆升.基于数据融合技术的路段出行时间预测方法[J].交通运输工程学报,2008(6):88-92.LIU Honghong,YANG Zhaosheng.Estimating methods of link travel times based on data fusion technology[J].Journal of Traffic and Transportation Engineering,2008(6):88-92.(in Chinese)
[10] 董春娇,邵春福,周雪梅,等.基于交通流参数相关的阻塞流短时预测卡尔曼滤波算法[J].东南大学学报(自然科学版),2014,44(2):413-419.DONG Chunjiao,SHAO Chunfu,ZHOU Xuemei,et al.Kalman filter algorithm for short-term jam traffic prediction based on traffic parameter correlation[J].Journal of Southeast University (Natural Science Edition).2014,44(2):413-419.(in Chinese)
[11] CHOI K,CHUNG Y.A data fusion algorithm for estimating link travel time[J].Journal of Intelligent Transportation Systems,2002,7(3):235-260.
[12] ZHENG W,LEE D H,ZHENG W,et al.Short-term freeway traffic flow prediction:Bayesian combined neural network approach[J].Journal of Transportation Engineering,2006,132(2):114-121.
[13] 王建,邓卫,赵金宝.基于改进型贝叶斯组合模型的短时交通流量预测[J].东南大学学报(自然科学版),2012,42(1):162-167.WANG Jian,DENG Wei,ZHAO Jinbao.Short-term freeway traffic flow prediction based on improved Bayesian combined model[J].Journal of Southeast University (Natural Science Edition),2012,42(1):162-167.(in Chinese)
[14] PARK D,RILETT L R,HAN G.Spectral basis neural networks for real-time travel time forecasting[J].Journal of Transportation Engineering,1999,125(125):515-523.
[15] 邱奉翠.基于城市快速路和主干道多源交通检测信息的数据融合技术研究[D].北京:北京交通大学,2012.QIU Fengcui.Study on data fusion technology of multi-source traffic data of urban expressway and arterial road[D].Beijing:Beijing Jiaotong University,2012.(in Chinese)
[16] 姚恩建,张永生.城市快速路区间旅行时间短期预测算法[J].武汉理工大学学报(交通科学与工程版),2013,37(6):1133-1137.YAO Enjian,ZHANG Yongsheng.A new algorithm of short-term travel time prediction for urban expressway[J].Journal of Wuhan University of Technology (Transportation Science & Engineering),2013,37(6):1133-1137.(in Chinese)
[17] 孙祥龙,陆建.公交站点间行程时间估计[J].武汉理工大学学报(交通科学与工程版),2013,37(1):67-70.SUN Xianglong,LU Jian.Travel time estimation between bus stop[J].Journal of Wuhan University of Technology (Transportation Science & Engineering),2013,37(1):67-70.(in Chinese)
[18] 黄靖翔.基于多源数据融合的高速公路路段行程时间估计[D].广州:华南理工大学,2018.HUANG Jingxiang.Link travel time estimation of expressway based on multi-source data fusion[D].Guangzhou:South China University of Technology,2018.(in Chinese)