Master's Thesis : Partially Detected Intelligent Traffic Signal Control using Connectionist Reinforcement Learning
Geortay, Cyril
Promotor(s) : Louppe, Gilles
Date of defense : 25-Jun-2020/26-Jun-2020 • Permalink : http://hdl.handle.net/2268.2/9068
Details
Title : | Master's Thesis : Partially Detected Intelligent Traffic Signal Control using Connectionist Reinforcement Learning |
Translated title : | [fr] Contrôle de feux de signalisation avec détection partielle par l'apprentissage par renforcement |
Author : | Geortay, Cyril |
Date of defense : | 25-Jun-2020/26-Jun-2020 |
Advisor(s) : | Louppe, Gilles |
Committee's member(s) : | Drion, Guillaume
Sabatelli, Matthia Geurts, Pierre |
Language : | English |
Number of pages : | 61 |
Keywords : | [en] Reinforcement Learning [en] Traffic light control |
Discipline(s) : | Engineering, computing & technology > Computer science |
Complementary URL : | https://github.com/CyrilGeo/PDITSCS |
Institution(s) : | Université de Liège, Liège, Belgique |
Degree: | Master en ingénieur civil en informatique, à finalité spécialisée en "intelligent systems" |
Faculty: | Master thesis of the Faculté des Sciences appliquées |
Abstract
[en] This master thesis focuses on traffic signal control using Reinforcement Learning with a neural network. It introduces an uncommon method by placing the technology necessary for state representation inside the vehicles. This prevents an expensive set up and maintenance of sensors at the traffic light intersection, but introduces a new problem: partial detection of the incoming vehicles.
File(s)
Document(s)
Annexe(s)
manhattan.png
Description: Topology of a Manhattan network used as a testing environment
Size: 23.31 kB
Format: image/png
Description: Topology of a Manhattan network used as a testing environment
Size: 23.31 kB
Format: image/png
flow_perf.png
Description: Performances in terms of waiting time of vehicles under different traffic flows and for different detection rates
Size: 31.28 kB
Format: image/png
Description: Performances in terms of waiting time of vehicles under different traffic flows and for different detection rates
Size: 31.28 kB
Format: image/png
hourly_real_w.png
Description: Performances over a day in terms of waiting time of vehicles on the deployment intersection for different detection rates
Size: 71.95 kB
Format: image/png
Description: Performances over a day in terms of waiting time of vehicles on the deployment intersection for different detection rates
Size: 71.95 kB
Format: image/png
Cite this master thesis
All documents available on MatheO are protected by copyright and subject to the usual rules for fair use.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.
The University of Liège does not guarantee the scientific quality of these students' works or the accuracy of all the information they contain.