A Hybrid Map-matching Algorithm for Real Time Passenger Information Systems via Mobile Phones and Crowd-Sourcing

The aim of any Real Time Passenger Information (RTPI) System is to provide accurate and efficient travel and transport information to users. Most of the existing RTPI systems are largely developed in urban areas; often, rural passengers are neglected due to lack of infrastructure (for example vehicle tracking system), less number of passengers, and problems with communication systems. In this research project, a passenger centric RTPI is proposed, which uses crowd-sourcing and mobile phones so that passengers are not only information consumers but are also providers of information to the system. In the proposed RTPI, passengers can allow the system to track their location - via their smart phones - when they are travelling on public transportation; this will compensate for the lack of vehicle tracking system in public transport. Map-matching (MM) algorithms integrate data from positioning sensors (GPS) with a digital GIS map in order to identify: firstly, the road link on which a vehicle is travelling; and secondly, to determine the vehicle?s location on that segment. In the proposed RTPI, at a given point of time, we receive a number of vehicle locations (latitude and longitude) from passengers travelling on a bus. In order to provide a precise vehicle location at a given point of time, a novel map-matching algorithm using fuzzy logic - which integrates multiple vehicle locations (obtained from passenger?s smart phones) with a GIS road map - has been developed. The developed map-matching algorithm has been tested using real-world data collected on four different bus routes in Aberdeenshire, Scotland. It was identified that the developed MM algorithm is efficient and capable of supporting the proposed passenger information system.