Please use this identifier to cite or link to this item:
Title: Safer Roads
Authors: Cremona, Francesco
Keywords: Mobile apps
Application software -- Development
Global Positioning System
Automobile driver education
Issue Date: 2018
Citation: Cremona, F. (2018). Safer Roads (Bachelor's dissertation).
Abstract: This project attempts to promote responsible driving and speed awareness through the development of a mobile application which provides users on speed limit information on roads on which they are travelling, together with a journey logging feature to later on review their driving performance. Tracked journeys can be displayed on a map and the trace shown as a heatmap, comparing the vehicle speed to the road speed limit. The aim is to promote responsible driving and road safety awareness. A comparative study of map matching algorithms is also conducted on the performance and accuracy of such algorithms, which match raw GPS data to the road network. Five di erent map matching algorithms are implemented and modi ed, taking into account parameters such as distance from a candidate road, direction difference, to- gether with other restriction and connectivity information. For testing, a dataset of various GPS traces in differing environments was developed and labelled, together with a trace visualisation tool which gives the ability to display, annotate and manage GPS datasets on a map. Accuracies were calculated for different building densities and GPS frequencies. The geometric algorithms tested were point to point matching with 68.2% accuracy, point to curve at 90% and two types of curve to curve algorithms at 90.7% and 91.2%. A weight-based topological algorithm achieved accuracies of 85.6% and 94.5% using different proposed urban and suburban weights, respectively. The most accurate matching was achieved using a modi ed version of this algorithm, with an accuracy of 94.9% at a GPS frequency of 1Hz. This algorithm together with three of the reviewed map matching algorithms were implemented on a mobile device and tested for real-time performance.
Appears in Collections:Dissertations - FacICT - 2018
Dissertations - FacICTCCE - 2018

Files in This Item:
File Description SizeFormat 
  Restricted Access
1.62 MBAdobe PDFView/Open Request a copy

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.