Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Traffic demand and longer term forecasting from real-time observations

Sopasakis, Alexandros LU (2019) 6th International Conference on Time Series and Forecasting p.1247-1259
Abstract
We optimize traffic signal timing sequences for a section of a traffic net-
work in order to reduce congestion based on anticipated demand. The system relies
on the accuracy of the predicted traffic demand in time and space which is carried
out by a neural network. Specifically, we design, train, and evaluate three different
neural network models and assert their capability to describe demand from traffic
cameras. To train these neural networks we create location specific time series data
by approximating vehicle densities from camera images. Each image passes through
a cascade of filtering methods and provides a traffic density estimate corresponding
to the camera location at that specific time. The system... (More)
We optimize traffic signal timing sequences for a section of a traffic net-
work in order to reduce congestion based on anticipated demand. The system relies
on the accuracy of the predicted traffic demand in time and space which is carried
out by a neural network. Specifically, we design, train, and evaluate three different
neural network models and assert their capability to describe demand from traffic
cameras. To train these neural networks we create location specific time series data
by approximating vehicle densities from camera images. Each image passes through
a cascade of filtering methods and provides a traffic density estimate corresponding
to the camera location at that specific time. The system is showcased using real-time
camera images from the traffic network of Goteborg. We specifically test this system
in reducing congestion for a small section of the traffic network. To facilitate the
learning and resulting prediction we collected images from cameras in that network
over a couple of months. We then use the neural network to produce forecasts of traffic
demand and adjust the traffic signals within that section. To simulate how congestion
will evolve once the traffic signals are adjusted we implement an advanced stochastic
model. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
traffic demnad, forecasting, lstm, gru, saes, image processing, time series
host publication
ITISE 2019 International Conference on Time Series and Forecasting : Proceedings of Papers 25-27 September 2019 Granada (Spain) - Proceedings of Papers 25-27 September 2019 Granada (Spain)
editor
Valenzuela, Olga ; Rojas, Fernando ; Pomares, Hector and Rojas, Ignacio
pages
1247 - 1259
conference name
6th International Conference on Time Series and Forecasting
conference location
Granada, Spain
conference dates
2019-09-25 - 2019-09-27
ISBN
978-84-17970-78-9
language
English
LU publication?
yes
id
9960c8d7-c05b-4693-a540-8fcdb2a674ea
date added to LUP
2019-10-10 21:28:40
date last changed
2022-12-08 14:06:17
@inproceedings{9960c8d7-c05b-4693-a540-8fcdb2a674ea,
  abstract     = {{We optimize traffic signal timing sequences for a section of a traffic net-<br/>work in order to reduce congestion based on anticipated demand. The system relies<br/>on the accuracy of the predicted traffic demand in time and space which is carried<br/>out by a neural network. Specifically, we design, train, and evaluate three different<br/>neural network models and assert their capability to describe demand from traffic<br/>cameras. To train these neural networks we create location specific time series data<br/>by approximating vehicle densities from camera images. Each image passes through<br/>a cascade of filtering methods and provides a traffic density estimate corresponding<br/>to the camera location at that specific time. The system is showcased using real-time<br/>camera images from the traffic network of Goteborg. We specifically test this system<br/>in reducing congestion for a small section of the traffic network. To facilitate the<br/>learning and resulting prediction we collected images from cameras in that network<br/>over a couple of months. We then use the neural network to produce forecasts of traffic<br/>demand and adjust the traffic signals within that section. To simulate how congestion<br/>will evolve once the traffic signals are adjusted we implement an advanced stochastic<br/>model.}},
  author       = {{Sopasakis, Alexandros}},
  booktitle    = {{ITISE 2019 International Conference on Time Series and Forecasting : Proceedings of Papers 25-27 September 2019 Granada (Spain)}},
  editor       = {{Valenzuela, Olga and Rojas, Fernando and Pomares, Hector and Rojas, Ignacio}},
  isbn         = {{978-84-17970-78-9}},
  keywords     = {{traffic demnad; forecasting; lstm; gru; saes; image processing; time series}},
  language     = {{eng}},
  month        = {{09}},
  pages        = {{1247--1259}},
  title        = {{Traffic demand and longer term forecasting from  real-time observations}},
  year         = {{2019}},
}