| 摘要: |
| 目前已有多种公交车到站时间预测技术,但在城市复杂多变的道路交通环境下预测精确度不够理想,为此提出了一种基于路况相似性的城市公交车到站时间预测机制,通过分析公交车运行历史数据获取各个路段的历史路况评估值,提取历史行程记录的路况特征.在运行阶段,实时获取公交车运行数据并评估各个路段的路况,查找与当前时间、位置和路况特征最相似的历史行程记录,并据此预测公交车辆的到站时间. |
| 关键词: 到站时间预测 路况特征 k 近邻 相似性 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金(60933011); 国家重点基础研究发展计划(973)(2011CB302902); 国家科技重大专项(2011ZX03005-006); 中国科学院先导专项课题(XDA06040100) |
|
| City Bus Arrival Time Prediction Based on Similarity of Road Conditions |
|
SUN Yu-Yan1,2,3, LIU Yan4, ZHOU Xin-Yun1, SUN Li-Min1
|
|
1.Institute of Information Engineering, The Chinese Academy of Sciences, Beijing 100093, China;2.Graduate University, The Chinese Academy of Sciences, Beijing 100049, China;3.State Key Laboratory of Information Security (Institute of Information Engineering, Th;4.School of Software and Microelectronics, Peking University, Beijing 102600, China
|
| Abstract: |
| Currently, bus arrival time prediction studies have not yielded the best results under complex road conditions. A bus arrival time prediction mechanism based on the similarity of road conditions is proposed in this paper. By analyzing the past trip records of the bus, the historical traffic condition of each road section was estimated. The road condition for each trip records was extracted. In the operational phase, each road section’s traffic condition was estimated by gathering bus locations continually. The mechanism predicted bus arrival time for the rest bus station by the most similar historical which was searched by the time, distance, and real-time road conditions. The evaluation result using historical data of real bus trips shows that the arrival time prediction of the mechanism is accurate—with an absolute percentage error around 5%. |
| Key words: arrival time prediction road condition signature k-nearest neighbor similarity |
