Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Train Transportation in Sri Lanka: A Smart IOT based Multi-Subsystem Approach using MQTT

Dhanushka Balasingham, Sadeesha Samarathunga, Anuththara Bandara, Gayantha Godakanda Arachchige, Narmada Gamage, Jaliya L. Wijayaraja

TL;DR

The paper addresses inefficiencies in Sri Lanka's historic railway by proposing an MQTT-driven, multi-subsystem ATMS that integrates GPS-based tracking, RFID-based e-ticketing, and vision-based passenger counting. The approach unifies these subsystems through a standardized MQTT topic schema and a centralized ATMS web platform, enabling real-time monitoring and decision support. Key findings include a GPS accuracy of about $24$ meters, MQTT-based location updates with average latency of $458.83$ ms versus $899.5$ ms for HTTP, and robust real-time counting demonstrated by high AUC-PR using YOLOv8; together, these results validate improved scheduling, passenger flow, and system management. The work demonstrates practical modernization and interoperability potential for Sri Lanka's railway infrastructure, offering a blueprint for scalable, vendor-agnostic deployment across modes of transportation.

Abstract

This research proposes a system as a solution for the challenges faced by Sri Lanka' s historic railway system, such as scheduling delays, overcrowding, manual ticketing, and management inefficiencies. It proposes a multi-subsystem approach, incorporating GPS tracking, RFID-based e-ticketing, seat reservation, and vision-based people counting. The GPS based real time train tracking system performs accurately within 24 meters, with the MQTT protocol showing twice the speed of the HTTP-based system. All subsystems use the MQTT protocol to enhance efficiency, reliability, and passenger experience. The study's data and methodology demonstrate the effectiveness of these innovations in improving scheduling, passenger flow, and overall system performance, offering promising solutions for modernizing Sri Lanka's railway infrastructure.

Enhancing Train Transportation in Sri Lanka: A Smart IOT based Multi-Subsystem Approach using MQTT

TL;DR

The paper addresses inefficiencies in Sri Lanka's historic railway by proposing an MQTT-driven, multi-subsystem ATMS that integrates GPS-based tracking, RFID-based e-ticketing, and vision-based passenger counting. The approach unifies these subsystems through a standardized MQTT topic schema and a centralized ATMS web platform, enabling real-time monitoring and decision support. Key findings include a GPS accuracy of about meters, MQTT-based location updates with average latency of ms versus ms for HTTP, and robust real-time counting demonstrated by high AUC-PR using YOLOv8; together, these results validate improved scheduling, passenger flow, and system management. The work demonstrates practical modernization and interoperability potential for Sri Lanka's railway infrastructure, offering a blueprint for scalable, vendor-agnostic deployment across modes of transportation.

Abstract

This research proposes a system as a solution for the challenges faced by Sri Lanka' s historic railway system, such as scheduling delays, overcrowding, manual ticketing, and management inefficiencies. It proposes a multi-subsystem approach, incorporating GPS tracking, RFID-based e-ticketing, seat reservation, and vision-based people counting. The GPS based real time train tracking system performs accurately within 24 meters, with the MQTT protocol showing twice the speed of the HTTP-based system. All subsystems use the MQTT protocol to enhance efficiency, reliability, and passenger experience. The study's data and methodology demonstrate the effectiveness of these innovations in improving scheduling, passenger flow, and overall system performance, offering promising solutions for modernizing Sri Lanka's railway infrastructure.
Paper Structure (13 sections, 4 equations, 5 figures)

This paper contains 13 sections, 4 equations, 5 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. GPS based real time train tracking system with alarm notification system
  5. RFID-based E-ticketing system
  6. An vision-based real time people counting system
  7. Communication
  8. ATMS
  9. Results and Discussion
  10. Evaluation of GPS tracking system
  11. Evaluation of HTTP and MQTT based GPS tracking system
  12. Evaluation of People counting system
  13. Conclusion

Figures (5)

  • Figure 1: System diagram of the ATMS.
  • Figure 2: Suggested MQTT topic for the transportation system.
  • Figure 3: Evaluation of location gathering using Neo 6m GPS module.
  • Figure 4: Performance of MQTT over HTTP protocol.
  • Figure 5: Evaluation of vision based people counting system.