Back to search

Aristotle University Of Thessaloniki (AUTH)

Γεωπληροφορική ανάλυση χωρο-χρονικών δεδομένων, για την ανάπτυξη έξυπνου συστήματος πλοήγησης οχημάτων σε αστικό ιστό

Abstract

dc:description

The past decade, the population of people living in urban areas has rapidly increased. The number of urban residents is continuously growing and is not expected to drop any time soon.Among other countries, Greece is a great example of this phenomenon. Considering that 78% of the population already lives in the largest cities of Greece. The exponential growth of urbanisation exacerbated traffic congestion in these cities due to their inadequate infrastructure. Some ramifications of this are the increase of vehicle emissions, which consequently caused environmental degradation, and travel time with its associated costs.A viable solution to reduce traffic congestion is the concept of "smart cities". An interesting component for research in this concept is the investigation of potential solutions that can alleviate traffic congestion with better traffic management. For instance, better traffic management could be achieved with the development of a traffic prediction system that would accurately forecast current traffic conditions. Based on the above context, this dissertation introduces a methodology that will be used for the development of a smart urban vehicle navigation system which predicts traffic conditions. These are the objectives of the proposed navigation system: (1) Development of a generic methodology beneficial for every urban road network. (2) Development of an automated smart system that will not require human intervention. (3) Development of a GIS-based system independent from every GIS software. (4) Development of a system that can produce an autonomous road network map. (5) Development of a tool that facilitates the navigation process, investigates possible paths aiming to find the optimum, predicts travel time and fuel emissions. (6) Development of a Graphical User Interface (GUI) that will be user friendly.The study area is the urban road network of Thessaloniki city. The required spatio-temporal data have been provided by taxi-vehicles which use GPS receivers to navigate in the city. While these taxi-vehicles travel along the road network of Thessaloniki following the general traffic flow, they provide a large amount of GPS traces that include vehicles' unique ID and coordinate, velocity, orientation, time and date. A data mining system is uniquely developed with the use of programming language for big data management. This spatio-temporal data mining system needs to fulfill strict selection criteria and is a very crucial part of the dissertation project is as it facilitates the data selection process with information that is only relevant to the study area.The selected data that are generated from this data mining system are then inserted to a map-matching process that results to more accurate data selection. In addition, this map-matching process establishes an important tool for the development of an autonomous road network map that can be adjustable to any change of the road network. A high accuracy system is also developed for a map quality assessment using the inertial system ELLIPSE N of the SBG company. The GPS/INS system collects spatio-temporal data that are used to calculate confidence intervals. Then, the spatial models of each road segment are evaluated.Travel time is calculated afterwards with the available spatio-temporal data, which are then used in the following algorithms to predict travel time: Kalman Filter, Autoregressive Integrated Moving Average (ARIMA), Artificial Neural Networks (ANN), k-nearest neighbor. After considering various statistic indicators, ANN is selected to be the optimum algorithm for this specific case-study.Part of the proposed methodology is used for the development of an application that navigates the drivers based on shortest travel time or the "green" path, which is a path with less vehicle emissions. The proposed methodology and the custom code attempt to integrate all three disciplines: Geoinformatics, spatial-temporal data analysis and optimum prediction algorithm. This integration constitutes an effective process to predict travel time for every urban area. The GIS-based application is user friendly, open source and has open architecture. This application can be used by every user that needs to make a decision, while, the results can be exported to files compatible with the GIS software.The current dissertation concludes that the spatio-temporal data from GPS receivers of taxi-vehicles include all the required information for the development of a navigation system based on travel time. A major advantage of the data analysis is that it can be implemented without using any GIS software, provided that appropriate coding has been developed. However, the use of a data mining system is a critical requirement for the development of the methodology. The implementation of a map-matching process is required for the final stage of the data selection process while it simultaneously leads to more accurate results. This makes the current dissertation to be advantageous as the system generates an autonomous road network map by itself. Hence, there is a "closed" system that helps to reduce accuracy issues and leads to an independent and adaptive system.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2017

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Kyriakou, Kalliopi
  • Κυριακού, Καλλιόπη

Subjects

dc:subject × 10

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/42352

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Kyriakou, Kalliopi; Κυριακού, Καλλιόπη. Γεωπληροφορική ανάλυση χωρο-χρονικών δεδομένων, για την ανάπτυξη έξυπνου συστήματος πλοήγησης οχημάτων σε αστικό ιστό. Aristotle University Of Thessaloniki (AUTH), 2017. http://hdl.handle.net/10442/hedi/42352