Sprungmarken

Servicenavigation

Hauptnavigation

Sie sind hier:

Hauptinhalt

Detailseite



Passenger Flows: Crowd Mobility Analytics with Edge Computing in Public Transport

Lehrstuhl: Lehrstuhl für Kommunikationsnetze

Betreuer: Fang-Jing Wu and Ralf Burda ,

Beginn ab: SS 2018

Maximale Anzahl der Teilnehmer: 6

Beschreibung: The goal of this project group is to exploit sensors integrated with Internet-of-Things (IoT) devices for monitoring passenger flows in public transport. This project group will address sensing, data analytics, and data visualization issues. Multi-modal sensors including GPS, Inertial Measurement Units (IMU), and Wi-Fi antennas will be integrated with lightweight Edge devices to perceive human mobility and environmental conditions. The collected sensing data will be analyzed on the lightweight devices. Data analytics algorithms will be designed to detect passengers in the train/bus and track the passenger mobility at train/bus stations. Finally, the data analytics results will be reported to the backend and will be visualized through a web service for end-users and stakeholders. An initial prototype system will be developed to conduct experiments with the H-Bahn Dortmund in the TU Dortmund campus for the first trial.

To achieve this goal, this project group will focus on the 3 technical sub-tasks: (1) sensing, (2) analytics, and (3) visualization.
• Sensing: The Wi-Fi sniffing technology with privacy protection mechanisms will be implemented on edge nodes for collecting human mobility data. Meanwhile, a GPS and an IMU sensor are integrated with the edge nodes to trigger passenger detection and update passenger flows.
• Analytics: On-board stop detection, passenger detection, and flow tracking algorithms will be designed to optimize the service delays and network usage.
• Visualization: The data analytics results will be represented through web services or mobile applications.

Programming languages for major tasks (but not limited to the following options):
• Sensing: Python or C
• Data analytics: Python, Java, or others
• Visualization: Java, Java script, or others


The 1st meeting is scheduled at 11:30 am on 9 April. Please contact Dr. Fang-Jing Wu for further information.