Inhalt des Dokuments
Es gibt keine deutsche Übersetzung dieser Webseite.
Master Thesis: Smartphone-based Public Transport Vehicle Identification Using GTFS
Smartphone-based Public Transport Vehicle Identification Using GTFS
A number of studies have shown that availability of real-time information on vehicle loca- tions and possible delays in public transportation have a positive impact on the passengers’ experience, and thus leads to improvement in urban mobility in general. With the ubiquity of smartphones, real-time public transportation information can be accessed anywhere and any- time. Contrary to the construction and operation of an infrastructure with electronic display boards at stops, the pervasiveness of smartphones bears an opportunity such that they could be utilized to produce real-time public transport data along with enabling their users to consume it.
This thesis proposes the design and implementation of a system that uses satellite location data obtained through smartphones of public transport users and static public transport net- work information publically available in GTFS format to identify a current trip of a public transport passenger such that the smartphone location data could be used to extend the static timetable data to generate real-time information about the passenger’s trip. The proposed sys- tem is evaluated based on correct identification of 846 minutes of GPS data collected for public transportation trips in Berlin, with respect to transit network data from the VBB.
Promising results compared to the related work are obtained, where in average 93% of data recorded from busses are correctly identified for mode of transport and up to 59% are identified correctly to the respective trip. Similar results are achieved for data collected from trains. These results can be regarded all the more positive considering that the evaluation has been made in the inner city of Berlin, which poses challenges for position data accuracy and in terms of the high spatial and temporal density of the transit network.
Supervisor: Sebastian Zickau 
Type: Master Thesis
Duration: 6 months
10587 Berlin, Germany
Phone: +49 30 8353 58811
Fax: +49 30 8353 58409