Traffic data interpolation method of non-detection road link based on Kriging interpolation
-
摘要: 从交通流扩散的特点和人的先验知识出发, 提出采用Kriging插值法对路网中无检测器路段进行交通数据插补。基于交通数据空间相关性的特征, 对交通数据进行空间建模, 从而以空间距离作为度量基准对未知路段交通数据进行估计。利用南昌市浮动车系统中提取的路段行程速度作为试验数据, 进行了试验验证。研究结果表明: 在城市交通中各个典型时段行程速度的插补值标准差可以控制在8 km·h-1以内; 在针对路网形态差异较大的中心区和湖区分别进行的试验中, 行程速度的平均绝对误差都保持在2~5 km·h-1之间。可见, 该方法具有良好的时态和区域移植性。Abstract: From the diffused characteristic of traffic flow and prior knowledge, Kriging interpolation was adopted to interpolate the traffic data of non-detection road link. Based on the spatial correlation of traffic data, a spatial model of traffic data was built. The spatial distance was adopted as metric to estimate the unsampled traffic data of road link. The road link travel speeds of Nanchang's road network were used as experiment data, which were collected from urban floating car system, and the method was verified. Experiment result shows that the standard errors of speed interpolations are always lower than 8 km·h-1 in different urban traffic time periods. Downtown zone and lake zone have different road network structures, and their mean absolute errors of speed interpolations are 2-5 km·h-1.So the method has good temporal and regional portabilities.
-
表 1 样本数据分布的偏度和峰度
Table 1. Skewnesses and kurtosises of sample data distributions
统计量 中心区 湖区 早高峰 平峰 晚高峰 周末 早高峰 平峰 晚高峰 周末 峰度 0.093 0.125 0.254 0.524 0.431 0.643 -0.118 0.065 偏度 0.194 -0.054 0.459 0.651 -0.245 -0.288 -0.411 -0.203 
下载: 导出CSV
表 3 交叉检验结果
Table 3. Cross validation results
误差 中心区 湖区 早高峰 平峰 晚高峰 周末 早高峰 平峰 晚高峰 周末 平均绝对误差/ (km·h-1) 3.47 3.65 3.20 2.36 4.62 4.39 4.67 3.95 平均相对误差/% 13.35 15.12 12.45 8.56 13.09 11.99 14.14 10.40 均方根误差/ (km·h-1) 4.56 4.80 4.44 3.46 6.32 6.22 6.57 5.14
下载: 导出CSV
-
[1] SMITH B L, SCHERER W T, CONKLIN J H. Exploringi mputation techniques for missing datain transportation management systems[J]. Transportation Research Record, 2003 (1836): 132-142. [2] 孙玲, 刘浩, 牛树云. 考虑时空相关性的固定检测缺失数据重构算法[J]. 交通运输工程学报, 2010, 10 (5): 121-126. doi: 10.3969/j.issn.1671-1637.2010.05.021SUN Ling, LIU Hao, NIU Shu-yun. Reconstructive method of missing data for location-specific detector considering spatiotemporal relationship[J]. Journal of Traffic and Transportation Engineering, 2010, 10 (5): 121-126. (in Chinese) doi: 10.3969/j.issn.1671-1637.2010.05.021 [3] ZHONG Ming, LINGRAS P, SHARMA S. Estimation of missing traffic counts using factor, genetic, neural, and regression techniques[J]. Transportation Research Part C: Emerging Technologies, 2004, 12 (2): 139-166. doi: 10.1016/j.trc.2004.07.006 [4] WANG Yi-bing, PAPAGEORGIOU M. Real-time freeway traffic state esti mation based on extended Kal man filter: a general approach[J]. Transportation Research Part B: Methodological, 2005, 39 (2): 141-167. doi: 10.1016/j.trb.2004.03.003 [5] TOBLER W R. A computer movie simulating urban growth in the Detroit Region[J]. Economic Geography, 1970, 46 (S): 234-240. [6] ZOU Hai-xiang, YUE Yang, LI Qing-quan, et al. A spatial analysis approach for describing spatial pattern of urban traffic state[C]//IEEE. 13th International IEEE Annual Conference on Intelligent Transportation Systems. Madeira Island: IEEE, 2010: 557-562. [7] BRAXMEIER H, SCHMIDT V, SPODAREV E. Kriged road-traffic maps[C]//PILZ J. Interfacing Geostatistics and GIS. Berlin: Springer, 2009: 105-119. [8] HACKNEY J K, BERNARD M, BI NDRA S, et al. Predicting road system speeds using spatial structure variables and network characteristics[J]. Journal of Geographical Systems, 2007, 9 (4): 397-417. [9] WANG Xiao-kun, KOCKELMAN K M. Forecasting network data: spatial interpolation of traffic counts using Texas data[J]. Transportation Research Record, 2009 (2105): 100-108. [10] MIURA H. A study of travel ti me prediction using universal Kriging[J]. TOP, 2009, 18 (1): 257-270. [11] CRESSIE N. The origins of Kriging[J]. Mathematical Geology, 1990, 22 (3): 239-252. [12] VER HOEF J M, PETERSON E, THEOBALD D. Spatial statistical models that use flow and stream distance[J]. Environmental and Ecological Statistics, 2006, 13 (4): 449-464. [13] YUE Yang, ZOU Hai-xiang, LI Qing-quan. Urban road travel speed esti mation based on low sampling floating car data[C]//WANG Yin-hai, YI Ping, AN Shi, et al. Proceedings of the 9thInternational Conference of Chinese Transportation Professionals. Harbin: ASCE, 2009: 1719-1725. -
点击查看大图
图(7) / 表(3)
计量
- 文章访问数: 837
- HTML全文浏览量: 208
- PDF下载量: 761
- 被引次数: 0
