Mobile

Crowdsourced Public Transport Arrival Tracker

Arrival predictions for transport networks that publish no live data, built from anonymous location traces contributed by passengers already on board. The interesting problems are estimating arrivals from sparse noisy traces and designing the privacy model.

Difficulty

Intermediate

A small team, or one strong student willing to learn something new.
Effort

1 semester, 2 to 3 students

Deliverables

5 to ship

3 optional extras

Suggested stack

React NativeNode.jsPostgreSQLPostGISRedis
A suggestion, not a requirement. Swap anything for what you already know.

What you should ship

  • Passenger mode contributing anonymised location traces while a journey is in progress
  • Route matching that snaps a trace to a known route and direction
  • Arrival estimation for downstream stops, with accuracy reported against observed arrivals
  • Rider view showing predictions with an explicit confidence and the age of the underlying data
  • Privacy design document covering anonymisation, retention and re-identification risk

If you have time left

  • Crowding reports from passengers alongside position
  • Historical schedule inference for routes with no published timetable
  • Graceful degradation to schedule based estimates when no live traces exist

The problem

Many transport networks, particularly informal or privately operated ones, publish no live data and often no timetable. Passengers wait without information, and the only people who know where a vehicle is are the people on it.

What you build

A contribution mode, route matching, arrival estimation, and a rider interface that is honest about uncertainty.

The cold start problem, stated honestly

With no contributors there are no predictions. Address this directly: pick two or three routes, get your project group and some volunteers contributing, and evaluate there. A well measured result on three routes is a real result. Claiming city wide coverage is not.

The privacy problem, which is the serious one

Location traces are among the most sensitive data there is, and a trace from a bus stop near someone's home to a stop near their workplace identifies them regardless of what you call the identifier. Your report needs a genuine threat model: what you collect, how long you keep it, what you truncate at journey ends, and what an attacker with the database could learn.

How to evaluate it

Stand at a stop with a stopwatch and record actual arrivals against what the app predicted. Report mean absolute error by prediction horizon.

Scope warning

Do not attempt journey planning or multi-modal routing. Live arrivals for a small number of routes is the project.

Ideas and guidance, not finished projects

These are project ideas and scoping guidance, published free for students to use as a starting point. I do not build, write, or sell final-year projects, and I do not complete coursework for anyone. Take an idea, make it yours, and build it.

Not sure this one fits you?

I help students pick an idea that matches their skills and their deadline, then scope it down until it is finishable. Guidance only, never done for you.

Get guidance
Work with me

Stuck on something specific?

Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Every session is free; a few slots open each week.

Follow along

New writing, resources and project ideas land here first.