Hybrid LSTM Neural Network for Short-Term Traffic Flow Prediction

Yuelei Xiao,Yang Yin

doi:10.3390/info10030105

Yuelei Xiao, Yang Yin

Open Access

https://doi.org/10.3390/info10030105

Copy DOI

Abstract

The existing short-term traffic flow prediction models fail to provide precise prediction results and consider the impact of different traffic conditions on the prediction results in an actual traffic network. To solve these problems, a hybrid Long Short–Term Memory (LSTM) neural network is proposed, based on the LSTM model. Then, the structure and parameters of the hybrid LSTM neural network are optimized experimentally for different traffic conditions, and the final model is compared with the other typical models. It is found that the prediction error of the hybrid LSTM model is obviously less than those of the other models, but the running time of the hybrid LSTM model is only slightly longer than that of the LSTM model. Based on the hybrid LSTM model, the vehicle flows of each road section and intersection in the actual traffic network are further predicted. The results show that the maximum relative error between the actual and predictive vehicle flows of each road section is 1.03%, and the maximum relative error between the actual and predictive vehicle flows of each road intersection is 1.18%. Hence, the hybrid LSTM model is closer to the accuracy and real-time requirements of short-term traffic flow prediction, and suitable for different traffic conditions in the actual traffic network.

Highlights

With the continuous expansion of urban size, the scale of urban traffic network is growing, and the number of vehicles is increasing
A hybrid Long Short–Term Memory (LSTM) neural network is proposed based on the LSTM model in this paper
The structure and parameters of the hybrid LSTM neural network are optimized for the large traffic flow set and the small traffic flow set

Summary

Introduction

With the continuous expansion of urban size, the scale of urban traffic network is growing, and the number of vehicles is increasing. It is necessary to predict the short-term traffic flow of the traffic network precisely and guide the traffic based on the prediction results, so as to enhance people’s travel experience, alleviate the traffic congestion problem, and provide a good decision support for the government to carry out the planning and construction of public traffic infrastructure. One is the type of models based on mathematical and physical methods, the other is the type of models based on simulation technology, neural network, fuzzy control, and other modern scientific and technological methods. The former mainly includes historical average method, parametric regression model, Auto Regressive Integrated Moving Average (ARIMA) model, Kalman filtering model, exponential smoothing model, etc. The former mainly includes historical average method, parametric regression model, Auto Regressive Integrated Moving Average (ARIMA) model, Kalman filtering model, exponential smoothing model, etc. [1,2,3,4,5,6,7] (see Section 2)

Objectives

Results

Conclusion