Implementation of a Real Time Passenger Information System

Abstract

Intelligent Transportation Systems (ITS) are gaining recognition in developing countries like India. This paper describes the various components of our prototype implementation of a Real-time Passenger Information System (RTPIS) for a public transport system like a fleet of buses. Vehicle-mounted units, bus station units and a server located at the transport company premises comprise the system. The vehicle unit reports the current position of the vehicle to a central server periodically via General Packet Radio Service (GPRS). An Estimated Time of Arrival (ETA) algorithm running on the server predicts the arrival times of buses at their stops based on real-time observations of the buses' current Global Positioning System (GPS) coordinates. This information is displayed and announced to passengers at stops using station units, which periodically fetch the required ETA from the server via GPRS. Novel features of our prototype include: (a) a route creator utility which automatically creates new routes from scratch when a bus is driven along the new route, and (b) voice tagging of stops and points of interest along any route. Besides, the prototype provides: (i) web-based applications for passengers, providing useful information like a snapshot of present bus locations on the streets, and (ii) web-based analysis tools for the transport authority, providing information useful for fleet management, like number of trips undertaken by a specific bus. The prototype has been demonstrated in a campus environment, with four-wheelers and two-wheelers emulating buses. The automatic real-time passenger information system has the potential of making the public transport system an attractive alternative for city-dwellers, thereby contributing to fewer private vehicles on the road, leading to lower congestion levels and less pollution.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 32 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.