Bus Dynamic Travel Time Prediction: Using a Deep Feature Extraction Framework Based on RNN and DNN

Yuan Yuan,Yimin Wang,Vlon Jang,Zhaocheng He,Zhichao Cao,Changsheng Zhu,Chunfu Shao

doi:10.3390/electronics9111876

Abstract

Travel time data is an important factor for evaluating the performance of a public transport system. In terms of time and space within the nature of uncertainty, bus travel time is dynamic and flexible. Since the change of traffic status is periodic, contagious or even sudden, the changing mechanism of that is a hidden mode. Therefore, bus travel time prediction is a challenging problem in intelligent transportation system (ITS). Allowing for a large amount of traffic data can be collected at present but lack of precisely-conducting, it is still worth exploring how to extract feature sets that can accurately predict bus travel time from these data. Hence, a feature extraction framework based on the deep learning models were developed to reflect the state of bus travel time. First, the study introduced different historical stages of bus signaling time, taxi speed, the stop identity (ID) of spatial characteristics, and real-time possible arrival time, signified by fourteen spatiotemporal characteristic values. Then, an embedding network is proposed to leverage a wide and deep structure to mate the spatial and temporal data. In order to meet the temporal dependence requirements, an attention mechanism for a Recurrent Neural Network (RNN) was designed in this research in order to capture the temporal information. Finally, a Deep Neural Networks (DNN) was implemented in this research in order to achieve the dynamic bus travel time prediction. Two case studies of Guangzhou and Shenzhen were tested. The results showed that the performance of the algorithm was more efficient than that of the traditional machine-learning model and promoted by 4.82% compared to the deep neural network applied to the initial feature space. Moreover, the study visualized the weighted cost of attention on the bus’s travel time features during a certain running state. Therefore, the study demonstrated the proposed model enabled to understand the characteristic data of transit travel time with visualization.

Highlights

We evaluated our approach using a large number of buses and taxi Global Positioning System (GPS) data, as well as the bus Automatic Vehicle Monitoring (AVL) data collected by the Transport Department of Guangzhou and Shenzhen in the south of China, which are metropolises with populations of over 14.9 million people and 13.2 million people, respectively
This section describes the evaluation of the accuracy of our approach for this study based on six types of experiments with our proposed model compared to the existing models
It is of particular importance to develop a deep-seated architecture that fully reflects the characteristics of transit travel time

Summary

Introduction

Bus travel time prediction has three dependencies.(1) Time [3]: Due to the strongBus travel time prediction is an important component of an dependence intelligent transport system (ITS).periodicity ofprecise passenger demand, bus scheduling has a certainThe capturing of real-time travel information facilitates theperiodicity.choice of an optimal routebus by atravelwith unforeseen eventshistorical occurring, travel traffic managers adjust departure schedules The time traveler.depends on the tendency of recent times. (2)Spatial dependence: in real time to ensure the service quality of a system [1,2].the travel time of the sameof the travel time of a particular line is influenced by the current traffic state variables bus route in the same city is dynamic due to the nature of bus operation because of frequent traffic running line and by the traffic state variables of the entire bus line [4]. (3) Exogenous congestion, traffic accidents, and road construction. Bus travel time prediction has three dependencies. (1) Time [3]: Due to the strong. Bus travel time prediction is an important component of an dependence intelligent transport system (ITS). The capturing of real-time travel information facilitates theperiodicity. With unforeseen eventshistorical occurring, travel traffic managers adjust departure schedules The time traveler. Depends on the tendency of recent times. Spatial dependence: in real time to ensure the service quality of a system [1,2]. (3) Exogenous congestion, traffic accidents, and road construction. It is necessary to focus on a real-time dependence: Some exogenous variables, weather conditions, and emergencies may have a great and dynamic bus travel time prediction model in depth in order to further improve traffic efficiency.

Objectives

Methods

Results

Conclusion