Date of Award

2016

Publication Type

Master Thesis

Degree Name

M.Sc.

Department

Computer Science

Keywords

Connected Vehicles, DSRC, GNSS, GPS, Intelligent Transportation System, Urban Canyon

Supervisor

Jaekel, Arunita

Rights

info:eu-repo/semantics/openAccess

Abstract

Connected Vehicles are expected to provide a major improvement in road safety. By broadcasting Basic Safety Messages (BSM) using Dedicated Short Range Communications (DSRC) all connected vehicles will have situational awareness of other connected vehicles in the area near them, and capability to provide ample warning of impending collisions. These systems rely on highly accurate GPS location data. GPS by design expects a clear line of sight (LoS) to four or more satellites for accuracy. City roads are often surrounded by buildings. These structures create areas isolated from sky views. Intelligent Transportation System (ITS) researchers have called these areas “urban canyons”. Buildings may block and/or bounce satellite signals, which can cause receivers to ‘see’ these signals either directly, indirectly, or both direct and indirect signals at the same time—which is the so-called multipath problem. Driving test results have been published which demonstrate the challenge. ITS researchers have noticed that position data taken by on-board units (OBU’s) contain these anomalies. When analyzed, these plots show vehicles as if they were driving through buildings. This is not helpful in preventing collisions. I will show that there is a data based approach to identify when Global Navigational Satellite System (GNSS) receivers are identifying impossible position results. I will also show a method using other available CAN bus data to interpolate expected geographic location and eliminate sending erroneous position reports.

Download

Share

COinS