公交到站时间预测及换乘机制的研究

发布时间：2018-11-14 12:35

【摘要】：随着智慧化城市的蓬勃发展，智能交通正受到人们越来越多的关注，而公交到站时间预测也成为了近年来的研究热点。实时准确的预测公交到站时间不仅能帮助出行者选择更好的出行路线，而且还能为交通部门科学管理、合理调度提供依据。本文针对公交到站时间预测问题和换乘问题进行了研究，提出了公交到站时间动态预测模型和基于公交到站时间预测的公交换乘模型，具体的研究内容包括以下三个方面：首先，探究了影响公交到站时间的各种静态因素和动态因素，建立了基于多种静态因素和动态因素的公交到站时间动态预测模型。实验结果表明，相对于基准模型，动态预测模型有效提高了预测准确度。其次，为了提高动态预测模型的效率和准确度，本文提出了基于波动性的自适应预测模型，该模型对历史数据波动性的统计主要包括以下3个方面：1）不同日期不同时间段到站时间波动性；2）不同日期不同路段到站时间波动性；3）不同时间段不同路段到站时间波动性。模型基于相邻路段到站时间波动相似性的考虑，将预测路线分成不同的路段组合进行分段预测。实验结果显示，自适应模型成功减少了模型的计算时间，并进一步提高了预测准确度。最后，针对现有公交换乘模型实时性和可靠性较低问题，，本文提出基于公交到站时间预测的公交换乘模型，该模型更多地考虑了时间段对换乘等车时间及换乘方式的影响，从而增强了换乘模型的实时性和可靠性。
[Abstract]:With the vigorous development of intelligent cities, people pay more and more attention to intelligent transportation, and bus arrival time prediction has become a hot topic in recent years. The real-time and accurate prediction of bus arrival time can not only help travelers choose a better route, but also provide scientific management and reasonable scheduling basis for traffic departments. In this paper, the problem of bus arrival time prediction and transit transfer is studied, and a dynamic prediction model of bus arrival time and a bus transfer model based on bus arrival time prediction are proposed. The specific research contents include the following three aspects: firstly, the static and dynamic factors affecting bus arrival time are explored, and a dynamic prediction model of bus arrival time based on various static and dynamic factors is established. The experimental results show that the dynamic prediction model can effectively improve the prediction accuracy compared with the benchmark model. Secondly, in order to improve the efficiency and accuracy of the dynamic prediction model, an adaptive prediction model based on volatility is proposed in this paper. The statistics of historical data volatility in this model mainly include the following three aspects: 1) the arrival time volatility of different dates and different time periods; 2) fluctuation of arrival time of different sections on different dates, 3) fluctuation of arrival time of different sections in different time periods. Based on the similarity of arrival time fluctuation of adjacent sections, the model divides the predicted routes into different combinations of sections for segmental prediction. The experimental results show that the adaptive model can reduce the computation time and improve the prediction accuracy. Finally, aiming at the problem of low real-time and reliability of the existing bus transfer model, this paper proposes a bus transfer model based on bus arrival time prediction. The model takes more account of the influence of time period on the transfer time and transfer mode. Thus, the real-time and reliability of the transfer model are enhanced.
【学位授予单位】：苏州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U491.17

【参考文献】