|
|
- 2017
利用GPS轨迹二次聚类方法进行道路拥堵精细化识别
|
Abstract:
针对当前在精细识别道路拥堵时空范围方面研究的不足，提出一种利用GPS轨迹的二次聚类方法，通过快速识别大批量在时间、空间上差异较小且速度相近的轨迹段，反映出道路交通状态及时空变化趋势，并根据速度阈值确定拥堵状态及精细时空范围。首先将轨迹按采样间隔划分成若干条子轨迹，针对子轨迹段提出相似队列的概念，并设计了基于密度的空间聚类的相似队列提取方法，通过初次聚类合并相似子轨迹段，再利用改进的欧氏空间相似度度量函数计算相似队列间的时空距离，最后以相似队列为基本单元，基于模糊C均值聚类的方法进行二次聚类，根据聚类的结果进行交通流状态的识别和划分。以广州市主干路真实出租车GPS轨迹数据为例，对该方法进行验证。实验结果表明，该二次聚类方法能够较为精细地反映城市道路的拥堵时空范围，便于管理者精准疏散城市道路拥堵，相比直接聚类方法可以有效提升大批量轨迹数据的计算效率
| [1] | Xia Ying, Zhang Xu, Wang Guoyin. Cluster-based Congestion Outlier Detection Method on Trajectory Data[C]. The 6th International Conference on Fuzzy Systems and Knowledge Discovery, Tianjin, China, 2009 |
| [2] | Song S, Kwak D, Kwak Y, et al. Segmentation Based Trajectory Clustering in Road Network with Location Sensing Technology[J].<em>Sensor Letters,</em> 2013, 11(9):1779-1782 |
| [3] | Wang Y, Han Q, Pan H. A Clustering Scheme for Trajectories in Road Networks[M].Berlin Heidelberg, Germany:Springer, 2012 |
| [4] | Ester M, Kriegel H P, Sander J, et al. A Density-based Algorithm for Discovering Clusters in Large Spatial Databases with Noise[C]. International Conference Knowledge Discovery and Data Mining, Portland, USA, 1996 |
| [5] | CJ J37-90. Specification for Design of Urban Road Subgrades[S]. Beijing, China:Ministry of Housing and Urban-Rural Development of the People's Republic of China, 1991(CJ J37-90. 城市道路设计规范[S]. 北京:中华人民共和国建设部, 1991) |
| [6] | Elnekave S, Last M, Maimon O. Incremental Clustering of Mobile Objects:Data Engineering Workshop[C].International Conference on Data Engineering Workshop, Istanbul, Turkey, 2007 |
| [7] | Li Qingquan, Hu Bo, Yue Yang. Flowing Car Data Map-Matching Based on Constrained Shortest Path Algorithm[J]. <em>Geomatics and Information Science of Wuhan University</em>, 2013, 38(7):805-808(李清泉,胡波,乐阳. 一种基于约束的最短路径低频浮动车数据地图匹配算法[J]. 武汉大学学报·信息科学版, 2013, 38(7):805-808) |
| [8] | Wang Handong, Yue Yang, Li Qingquan. How Many Probe Vehicles Are Enough for Identifying Traffic Congestion? -A Study from a Streaming Data Perspective[J]. <em>Frontiers of Earth Science</em>, 2013, 7(1):34-42 |
| [9] | Zhao Hongbin, Han Qilong, Pan Haiwei. Spatio-Temporal Similarity Measure for Trajectories on Road Networks[J]. <em>Computer Engineering and Applications</em>,2010, 46(29):9-12 |
| [10] | Ma Linbing, Li Peng. Spatio-Temporal Trajectory Clustering Based on Automatic Subspace Clustering Algorithm[J]. <em>Geography and Geo-Information Science</em>, 2014,30(4):7-11(马林兵,李鹏. 基于子空间聚类算法的时空轨迹聚类[J]. 地理与地理信息科学. 2014,30(4):7-11) |
| [11] | Sun Jian, Liu Qiong, Peng Zhongren. Research and Analysis on Causality and Spatial-Temporal Evolution of Urban Traffic Congestions-A Case Study on Shenzhen of China[J]. <em>Journal of Transportation Systems Engineering and Information Technology</em>, 2011,11(5):86-93(孙健,刘琼,彭仲仁. 城市交通拥挤成因及时空演化规律分析——以深圳市为例[J]. 交通运输系统工程与信息. 2011,11(5):86-93) |
| [12] | Wang Jingyuan, Mao Yu, Li Jing, et al. Predictability of Road Traffic and Congestion in Urban Areas[J]. <em>Plos One,</em>2015, 10(4):e01218254 |
| [13] | Chengkun L, Kun Q, Chaogui K. Exploring Time-dependent Traffic Congestion Patterns from Taxi Trajectory Data[C]. IEEE International Conference on Spatial Data Mining and Geographical Knowledge Services, Fuzhou, China, 2015 |
