Ανάλυση οδηγικής συμπεριφοράς με τη χρήση αλγορίθμων μηχανικής μάθησης
View/
Abstract
Worldwide, the increasing number of road accidents is one of the most serious issues and has an impact, not only on the state economy but also on the society itself. Nowadays, prediction and improvement of driving behaviour are keys to contribute to road safety by using machine learning algorithms. The taken information about driving behaviour can be important data for automotive industries, insurance companies and other new applications.
In recent years, due to the increase in computing power and computer systems storage spaces, significant growth has happened in the scientific field of machine learning. The use of machine learning enables computers to make predictions using experimental data without the need for human intervention beforehand. The use of these modern and innovative methods tends to go beyond the narrow confines of industrial and professional processes. Now with the use of Internet of Things (IoT) and Internet of Vehicles (IoV) it is used equally in daily activities such as driving a vehicle.
The dissertation aims to examine and distinguish the different ways of driving using machine learning algorithms. This research includes an approach by recognizing the driving behaviour of different drivers and a variety of routes.
The time-series data has generated by vehicle sensors. Specifically, in the present study, various clustering techniques applied to the data set.
In the end, in the research results, I will attempt to assess the behavioural driving of a set of drivers using collected data, while driving on various routes. I will group their data using the clustering method and then categorize them according to the driving style.
Postgraduate Studies Programme
Ψηφιακά Συστήματα και ΥπηρεσίεςDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
55Language
GreekCollections
Except where otherwise noted, this item's license is described as
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές