Campus Route Planner

Context

Built for a university software engineering module, this project models campus paths as a weighted graph and lets users optimise for more than raw distance.

What I wanted to build

A route planner that could explain why a path was chosen rather than returning only a line on a map.

System design

Path segments carry attributes for gradient, shelter, accessibility, and operating hours. The routing engine turns a selected preference into edge weights before running Dijkstra’s algorithm.

Hardware and software

The project uses TypeScript, Astro, GeoJSON, and a graph-based routing engine.

Build process

The team validated routes against real walks and added clear explanations for why a route was selected. That made the algorithm easier to debug and the result easier to trust.

Problems encountered

Different route preferences could produce technically correct paths that still felt surprising to a person walking them.

How I solved them

We compared routes against real walks and surfaced the weighting factors in the interface.

Final result

The prototype can compare fastest, sheltered, and accessible route priorities.

What I learned

An algorithmic result becomes more useful when the system can explain its trade-offs.

What I would improve next

I would improve operating-hours data and test the accessibility weights with more users.