摘要
使用出租车GPS数据作为基础,采用更加合理有效的路段速度作为交通状态参数,分析路网划分后的路段速度时间序列,利用四分位数特性优化算法,提高预测模型的合理性和准确性,并通过真实历史数据验证方法的可靠性.从带有随机性和不确定性的交通流变化中,通过分析找出其中的规律性,以预测未来几个时段的交通流变化.结果表明四分位法既体现出了路段速度的变化趋势,同时削弱了极端值和异常值的影响,能够展现出合理的交通状态变化过程,并且其计算简便,为大规模数据处理有效节省了计算资源.对计算结果的曲线拟合证明了四分位法处理路段速度的可靠性,对交通状态预测具有重要意义.
Taking taxi GPS data as basis and logical and valid road-segment speed as traffic state parameter,the time series of road-segment speed on meshed road net is analyzed and the tetra-quantile characteristic optimization algorithm is used to improve the rationality and accuracy of the prediction model and verify the reliability of this method with real historical data.By means of analysis,the regularity of traffic flow variation is found from its randomness and uncertainty to predict it in oncoming couple of time-intervals.The result shows that the tetra-quantile method will embody the trend of road-segment speed as well as weaken the influence of its extreme and abnormal values,showing a reasonable traffic state variation process.Besides,its calculation will be simple,saving computing resources effectively for large-scale data processing.The fitted road-segment speed curves will successfully verify that the road-segment processing with tetra-quantile method is reliable and significant for traffic state prediction.
引文
[1]AGBOLOSU-AMISON S J,PARK B,YUN I.Comparative evaluation of heuristic optimization methods in urban arterial network optimization[C]//12th International IEEE Conference on Intelligent Transportation Systems.[S.l.]:IEEE,2009:1-6.
[2]BAN X,HERRING R,HAO P,et al.Delay pattern estimation for signalized intersections using sampled travel times[C]//Proceedings of the 88th Annual Meeting of the Transportation Research Board.Washington:[s.n.],2009.
[3]FABRITIIS C,RAGONA R,VALENTI G.Traffic estimation and prediction based on real time floating car data[C]//11th International IEEE Conference on Intelligent Transportation systems.[S.l.]:IEEE,2008:197-203.
[4]HERRING R,HOFIEITNER A,AMIN S,et al.Using mobile phones to forecast arterial traffic through statistical learning[C]//89th Annual Meeting Transportation Research Board.Washington:[s.n.],2010.
[5]HUNTER T,HERRING R,ABBEEL P,et al.Path and travel time inference from GPS probe vehicle data[J/OL].[2016-06-20].https://www.researchgate.net/publication/229034951_Path_and_Travel_Time_Inference_from_GPS_Probe_Vehicle_Data.
[6]WORK D,BLANDIN S,TOSSAVAINEN O P,et al.A distributed highway velocity model for traffic state reconstruction[J/OL].[2016-06-20].http://piccoli.camden.rutgers.edu/files/IEEE_TAC27.pdf.
[7]YIM Y B Y,CAYFORD R.Investigation of vehicles as probes using global positioning system and cellular phone tracking:field operational test[R].California:Institute of Transportation Studies,University of California,2001.
[8]张红,朱昶胜.基于大数据的智能交通体系架构[J].兰州理工大学学报,2015(2):112-115.
[9]杨涛.基于浮动车技术的路段交通流量推算研究[D].北京:北京交通大学,2006.
[10]张周强,杜豫川.国外浮动车技术发展及其数据利用综述[C]//第七届华人交通运输学术大会论文集.上海:上海人民交通出版社,2007.
[11]姚智胜.基于实时数据的道路网短时交通流预测理论与方法研究[D].北京:北京交通大学,2007.
[12]朱丽云,温慧敏,孙建平.北京市浮动车交通状况信息实时计算系统[J].城市交通,2008,6(1):77-80.
[13]王晋生,周伟刚,王渤.GPS浮动车数据在城市交通管理和出行服务中的应用[J].公安交通科技窗,2009(1):31-33.
[14]董均宇.基于GPS浮动车的城市路段平均速度估计技术研究[D].重庆:重庆大学,2006.
[15]翁剑成,荣建,于泉.基于浮动车数据的行程速度估计算法及优化[J].北京工业大学学报,2007,33(5):459-464.