基于动态时间规整的出租车终点和行程时间预测
详细信息 查看官网全文
- 英文论文题名：The Destination and Travel Time of Taxi Trips Prediction Based on Dynamic Time Warping
- 论文作者：蔡恒兴 ; 栗波
- 英文论文作者：Cai Hengxing ; Li Bo ; Research Center of Intelligent Transportation System ; Sun Yat-sen University//Guangdong Provincial Key Laboratory of Intelligent Transportation System
- 年：2016
- 作者机构：中山大学智能交通研究中心//广东省智能交通系统重点实验室;
- 论文关键词：球面半正矢公式 ; 动态时间规整 ; 特征工程 ; 机器学习
- 英文论文关键词：Haversine Formula ; Dynamic Time Warping ; Feature Engineering ; Machine Learning
- 会议召开时间：2016-11-16
- 会议录名称：第十一届中国智能交通年会大会论文集
- 语种：中文
- 分类号：U492.434
- 学会代码：ZJTX
- 会议名称：第十一届中国智能交通年会
- 会议地点：中国重庆
- 主办单位：中国智能交通协会
- 学会名称：中国智能交通协会
- 页数：8
- 文件大小：1183k
- 原文格式：D
- 会议级别：全国
摘要
出租车是大众交通出行方式中的重要组成部分,高效的出租车调度有助于减少车辆空驶及缩短等待时间,从而节约运营成本和道路资源。本文旨在利用出租车的历史出行轨迹和当前已知的部分行驶轨迹,预测出租车出行终点和行程时间,为出租车管理者动态调度车辆资源提供决策支持。采用了球面半正矢公式和动态时间弯曲距离算法,在历史数据集中进行坐标和时间的特定查找,并结合特征工程和机器学习的方法。实验测试表明,本算法具有较高的预测精度。
Taxi plays an important role in public transportation,efficient vehicle dispatch helps to save operating costs and road resources by reducing deadhead and waiting time.There are two goals in this paper.Firstly,utilizing taxis' historical travel trajectories and parts of the currently known traveling trajectories to predict their destination of the traveling and corresponding travel time.Secondly,providing a decision support for operators to manage dynamic scheduling resources of taxi.Using Haversine formulas and dynamic time warping distance algorithm to find a particular character on the coordinates and time in the historical data.Using Feature Engineering and machine learning methods for prediction.The experimental test shows that the algorithm has a higher prediction accuracy.
Taxi plays an important role in public transportation,efficient vehicle dispatch helps to save operating costs and road resources by reducing deadhead and waiting time.There are two goals in this paper.Firstly,utilizing taxis' historical travel trajectories and parts of the currently known traveling trajectories to predict their destination of the traveling and corresponding travel time.Secondly,providing a decision support for operators to manage dynamic scheduling resources of taxi.Using Haversine formulas and dynamic time warping distance algorithm to find a particular character on the coordinates and time in the historical data.Using Feature Engineering and machine learning methods for prediction.The experimental test shows that the algorithm has a higher prediction accuracy.
引文
[1]Xuesong Zhou,Hani S.Mahmassani.A structural state space model for real-time traffic origin–destination demand estimation and prediction in a day-to-day learning framework[J].Transportation Research Part B,2007(41):823–840.
[2]Mahmassani,H.S..Dynamic network traffic assignment and simulation methodology for advanced system management applications[J].Networks and Spatial Economics,2001,1(2),267–292.
[3]Xuesong Zhou,et al.Estimating the most likely space–time paths,dwell times and path uncertainties from vehicle trajectory data:A time geographic method[J].Transportation Research Part C,2016(66):176–194.
[4]Martin L.Hazelton.Inference for origin-destination matrices estimation,prediction and reconstruction[J].Transportation Research Part B,2001(35):667-676.
[5]Robusto C C.The cosine-haversine formula[J].The American Mathematical Monthly,1957,64(1):38-40.
[6]Keogh E,Ratanamahatana C A.Exact indexing of dynamic time warping[J].Knowledge and information systems,2005,7(3):358-386.
[7]Turner,C.Reid,et al.A conceptual basis for feature engineering[J].Journal of Systems and Software,1999,49(1):3-15.
[8]Meinard Müller.Information Retrieval for Music and Motion[M].Springer Berlin Heidelberg,2007.
[2]Mahmassani,H.S..Dynamic network traffic assignment and simulation methodology for advanced system management applications[J].Networks and Spatial Economics,2001,1(2),267–292.
[3]Xuesong Zhou,et al.Estimating the most likely space–time paths,dwell times and path uncertainties from vehicle trajectory data:A time geographic method[J].Transportation Research Part C,2016(66):176–194.
[4]Martin L.Hazelton.Inference for origin-destination matrices estimation,prediction and reconstruction[J].Transportation Research Part B,2001(35):667-676.
[5]Robusto C C.The cosine-haversine formula[J].The American Mathematical Monthly,1957,64(1):38-40.
[6]Keogh E,Ratanamahatana C A.Exact indexing of dynamic time warping[J].Knowledge and information systems,2005,7(3):358-386.
[7]Turner,C.Reid,et al.A conceptual basis for feature engineering[J].Journal of Systems and Software,1999,49(1):3-15.
[8]Meinard Müller.Information Retrieval for Music and Motion[M].Springer Berlin Heidelberg,2007.