AAPpaper19102019 v1.pdf (1.13 MB)
Predicting real-time traffic conflicts using deep learning
journal contribution
posted on 2020-02-24, 14:52 authored by Nicolette Formosa, Mohammed Quddus, Stephen Ison, Mohamed Abdel-Aty, Jinghui YuanRecently, technologies for predicting traffic conflicts in real-time have been gaining momentum due to their proactive nature of application and the growing implementation of ADAS technology in intelligent vehicles. In ADAS, machine learning classifiers are utilised to predict potential traffic conflicts by analysing data from in-vehicle sensors. In most cases, a condition is classified as a traffic conflict when a safety surrogate (e.g. time-to-collision, TTC) crosses a pre-defined threshold. This approach, however, largely ignores other factors that influence traffic conflicts such as speed variance, traffic density, speed and weather conditions. Considering all these factors in detecting traffic conflicts is rather complex as it requires an integration and mining of heterodox data, the unavailability of traffic conflicts and conflict prediction models capable of extracting meaningful and accurate information in a timely manner. In addition, the model has to effectively handle large imbalanced data. To overcome these limitations, this paper presents a centralised digital architecture and employs a Deep Learning methodology to predict traffic conflicts. Highly disaggregated traffic data and in-vehicle sensors data from an instrumented vehicle are collected from a section of the UK M1 motorway to build the model. Traffic conflicts are identified by a Regional–Convolution Neural Network (R-CNN) model which detects lane markings and tracks vehicles from images captured by a single front-facing camera. This data is then integrated with traffic variables and calculated safety surrogate measures (SSMs) via a centralised digital architecture to develop a series of Deep Neural Network (DNN) models to predict these traffic conflicts. The results indicate that TTC, as expected, varies by speed, weather and traffic density and the best DNN model provides an accuracy of 94% making it reliable to employ in ADAS technology as proactive safety management strategies. Furthermore, by exchanging this traffic conflict awareness data, connected vehicles (CVs) can mitigate the risk of traffic collisions.
History
School
- Architecture, Building and Civil Engineering
Published in
Accident Analysis and PreventionVolume
136Issue
March 2020Publisher
ElsevierVersion
- AM (Accepted Manuscript)
Rights holder
© Elsevier LtdPublisher statement
This paper was accepted for publication in the journal Accident Analysis and Prevention and the definitive published version is available at https://doi.org/10.1016/j.aap.2019.105429.Acceptance date
2019-12-29Publication date
2020-01-10Copyright date
2020ISSN
0001-4575Publisher version
Language
- en
Depositor
Prof Mohammed Quddus. Deposit date: 20 February 2020Article number
105429Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC