Using Modern Vehicles as Mobile Sensors for Intelligent Traffic Awareness
Doctoral thesis
Permanent lenke
https://hdl.handle.net/11250/2982897Utgivelsesdato
2022Metadata
Vis full innførselSamlinger
Sammendrag
Traffic management has become a critical problem with growing traffic congestion worldwide. As a result, the approaches to managing traffic tend to become smart, and Intelligent Traffic Management Systems (ITMSs) are becoming increasingly common. Accessing the traffic data is a key component in ITMSs. As vehicles and their sensing and connection capabilities are advancing, more studies are investigating the feasibility of using Modern Vehicles (MVs) in estimating and sharing traffic data. However, despite the identified potential and increasing attention given to ITMSs and MVs, only a few studies use an MV as a mobile sensor to collect traffic data, with the purpose of improving the ITMS’s performance in a metropolitan area.
To address this gap, this thesis first identifies, by reviewing the literature, the existing methodologies and the corresponding traffic data types required by ITMSs and explores the potential research gaps in the field. Second, it addresses how a single MV and mounted low-cost sensors (i.e., a monocular camera with a built-in Global Positioning System [GPS] receiver) can be employed to estimate the traffic data of both MVs (i.e., geolocation) and the surrounding observed vehicles (i.e., number, type, relative position, distance, speed, lane, and geolocation) in a metropolitan area with a combination of Human-Driven Vehicle (HDV) and MV traffic. Third, it explores how the estimation error of a sensor mounted on a single MV can be mitigated to provide a more accurate picture of the traffic scene than what can be obtained by using data from only one MV, by fusing estimated traffic data (i.e., HDV’s geolocations) of two observing MVs.
In the initial stage of this study, a Systematic Literature Review (SLR) is conducted. Then, the Design Science Research (DSR) methodology is applied. Case studies are performed to evaluate and validate the proposed approaches and algorithms in terms of acceptance, usability, and impact on the problem at stake.
This thesis aims to bring researchers to the forefront of this new interdisciplinary field. The thesis contributes new knowledge to both the Computer Science (CS) and Civil Engineering (CE) fields and guides the design and prototyping of traffic data estimation by MVs to generate a dynamic model of the traffic scene that can enhance the performance of ITMSs. The lessons learned by the author of this thesis provided knowledge about the feasibility of using MVs to enhance traffic awareness by generating a dynamic model of the traffic scene.
Består av
Paper 1: Namazi, Elnaz; Li, Jingyue; Lu, Chaoru. Intelligent intersection management systems considering autonomous vehicles: A systematic literature review. IEEE Access 2019 ;Volum 7. s. 91946-91965 © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Paper 2: Namazi, Elnaz; Holthe-Berg, Rein Nisja; Skar Lofsberg, Christoffer; Li, Jingyue. Using Vehicle-Mounted Camera to Collect Information for Managing Mixed Traffic. I: 2019 15th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS 2019). IEEE 2019 ISBN 978-1-7281-5686-6. s. 222-230. © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Paper 3: Namazi, Elnaz; Li, Jingyue; Mester, Rudolf; Lu, Chaoru. Identifying and Counting Vehicles in Multiple Lanes by Using a Low-Cost Vehicle-Mounted Sensor for Intelligent Traffic Management Systems. I: Hybrid Artificial Intelligent Systems, 15th International Conference, HAIS 2020. Springer 2020 ISBN 978-3-030-61704-2. s. 598-611.
Paper 4: Namazi, Elnaz; Mester, Rudolf; Lu, Chaoru; Log, Markus Metallinos; Li, Jingyue. Improving Vehicle Localization with Two Low-Cost GPS Receivers. I: Innovations in Smart Cities Applications. Springer 2021 ISBN 978-3-030-94191-8.
Paper 5: Namazi, Elnaz; Mester, Rudolf; Lu, Chaoru; Li, Jingyue. Geolocation estimation of target vehicles using neural network-based image processing and geometric computations. : Neurocomputing Journal - Elsevier 2021 15 s.
Paper 6: Namazi, Elnaz; Mester, Rudolf; Li, Jingyue; Lu, Chaoru; Tang, Meng; Xiong, Ying. Traffic Awareness Through Multiple Mobile Sensor Fusion. This paper is not yet published and is therefore not included.