Background
A University of Liverpool final-year project for indoor route recording and personalized navigation. The administration app records routes and manages graphs; the user app retrieves routes and provides real-time guidance.
Goal
Represent routes with reusable nodes and edges so local updates do not require recording the entire environment again, and provide visual real-time guidance.
My contributions
- Designed two iOS apps with separate administration and navigation responsibilities.
- Reconstructed walking trajectories using CoreMotion heading estimation and step detection.
- Modeled routes as nodes and edges to support modular route-segment updates.
- Integrated Firebase Firestore for routes and nodes, with ARKit navigation overlays.
- Used QR codes to associate the initial user position with predefined graph nodes.
Approach
- Converted walking motion into trajectories, then organized reusable route segments as nodes and edges.
- Separated route administration from live navigation to keep the two workflows clear.
- Combined QR-based initialization, CoreMotion motion data and ARKit visual guidance.
Results
- Integrated route recording, cloud management and real-time AR guidance across administration and user apps.
- Supported graph updates by route segment, avoiding full-environment re-recording for each update.
The resume provides no positioning error, participant count or navigation completion rate; no performance improvement is inferred.