Système et gestion des télécommunications

Mobile Computing Application In Advanced Urban Public Transportation System

Abstract

Manish Chaturvedi

The ICT based advanced urban public transportation systems (APTS) are used to improve the commuters’ experience during their trips using public transportation buses. Generally, the deployment of APTS solutions requires the manual route feedings which is a tedious task specifically when the route changes are frequent. We design an automatic bus-stop detector using the GPS traces of buses and the crowd sourced accelerometer data of the commuters. We leverage the stop-and-go pattern of buses observed during their trips. The historical trajectory data of buses are processed using the DBSCAN algorithm to find all the locations where a bus stops recurrently during its trip (including the non-bus-stop locations). To discard the non-bus-stop locations, we utilize the fact that the bus-stops are the locations where commuters board or alight the bus. When a commuter is near the location detected as a stoppage by the DBSCAN algorithm, the accelerometer data of the commuter are processed in her smart phones to detect the bus-boarding event. The process involves the stepping event detection and the transport mode classification. The locations of bus-boarding events are used to extract all the bus-stops on the route accurately.