Sefism is free for learners. We are here to help you grow, one step at a time.Join for free
Sefism.
Mobile

Indoor Navigation App Using Bluetooth Beacons and Sensor Fusion

An Android app that gives turn-by-turn navigation inside a hospital, university block or mall where GPS does not work, using low-cost Bluetooth beacons fused with step detection and compass heading. An advanced mobile final year project with a strong accessibility angle for blind and partially sighted users.

Topics
Difficulty

Advanced

Needs a team, real planning, and a supervisor who knows the area.
Effort

2 semesters · 2 to 3 students

Deliverables

7 to ship

4 optional extras

Suggested stack

KotlinAndroid BLE scanning APIsAndroid sensor APIsParticle or Kalman filterNode.jsPostgreSQLGeoJSON indoor map
A suggestion, not a requirement. Swap anything for what you already know.

What you should ship

  • Beacon deployment across one real floor of a real building, with surveyed positions and a written deployment guide
  • Positioning engine combining signal strength with step detection and heading in a particle or Kalman filter
  • Indoor map model in GeoJSON with walkable paths, doors, stairs and lifts, and a routing graph built from it
  • Turn-by-turn guidance with route recalculation when the user deviates
  • Audio-first guidance designed for blind and partially sighted users, tested with TalkBack on
  • Accuracy evaluation at marked ground truth points, reported as median and 90th percentile error, with and without sensor fusion
  • An admin tool or script to register beacons and update the map without changing app code

If you have time left

  • Accessible routing that avoids stairs for wheelchair users
  • Multi-floor transitions using the barometer to detect floor changes
  • Automatic calibration that reduces the manual survey effort for a new building
  • Comparison against Wi-Fi round-trip time positioning on phones that support it

The problem with finding your way indoors

Satellite positioning stops working indoors, which is exactly where people get lost: hospitals with numbered wings, airports, large university blocks and shopping centres. For most visitors this costs time and stress. For blind and partially sighted people, indoor wayfinding is a daily obstacle that often means depending on someone else. Bluetooth beacons are cheap and battery powered, but turning their noisy signals into reliable guidance is hard.

What already exists and the gap

  • NavCog, from Carnegie Mellon University and IBM Research, is a smartphone navigation assistant for blind users that uses Bluetooth beacons for turn-by-turn guidance, and its published evaluations are worth reading before you start.
  • Situm is a commercial indoor positioning and wayfinding platform using beacons, Wi-Fi and sensors, deployed in airports and hospitals.
  • Google Maps has indoor floor plans for some large venues, but no turn-by-turn positioning inside most buildings.

Commercial platforms are priced and sold for large venues, and research systems like NavCog show what is possible but are not something a local hospital or campus can deploy on its own. The gap is an open, documented setup that one building's facilities team can install on one floor with low-cost beacons, and an app designed for audio guidance from the start.

What you build

Three parts:

  • Deployment kit: beacons mounted at surveyed positions, a GeoJSON floor map, and a script or admin page to register them.
  • Android app: scans beacons, fuses them with phone sensors, shows position on the map, routes to a destination and speaks instructions.
  • Small backend: serves maps and beacon registries per building, and collects anonymous accuracy logs during testing.

Main flow: pick a destination by search or voice, get a route, walk while the app updates position and announces turns, recalculate when the user goes the wrong way.

The technically interesting part

Sensor fusion. Bluetooth signal strength is a poor distance estimate. Walls, bodies, phone orientation and interference make trilateration jump by several metres. The project becomes interesting when you fuse beacon readings with the accelerometer and compass: detect steps, estimate heading, and use a particle filter where particles move with each step, are constrained by walls in the map, and are reweighted by beacon observations. Show your supervisor the position trace with and without fusion on the same walk. The improvement is your headline finding.

Guidance that is safe to follow blind. Instructions must be timely and conservative: announce a turn early enough, confirm after it, and say when the app is unsure rather than guessing.

How to evaluate it

  • Accuracy: mark 20 or more ground truth points on the floor; walk fixed routes and report median and 90th percentile error with beacons only and with fusion.
  • Guidance success: have 8 to 10 participants reach three destinations; record success rate, time and wrong turns.
  • Accessibility: if possible, test with blind or partially sighted volunteers through a local organisation, with consent and a sighted escort; otherwise, blindfolded sighted testers with an escort, reported as a limitation.
  • Battery: phone battery use per 30 minutes of navigation.

Suggested plan

  1. Weeks 1 to 4: choose the building, get permission, survey the floor, draw the GeoJSON map.
  2. Weeks 5 to 8: beacon scanning, raw trilateration, first accuracy baseline.
  3. Weeks 9 to 14: step detection, heading, particle filter, map constraints.
  4. Weeks 15 to 20: routing, turn-by-turn, audio guidance, recalculation.
  5. Weeks 21 to 26: user tests, accuracy study, deployment guide, write-up.

Path to a product

Your first customer is the building you tested in: a hospital, a university administration or a mall management office, offered as a pilot for one floor. Charge a setup fee per floor plus a small yearly maintenance fee covering beacon batteries and map updates. Disability organisations are natural partners and can help reach buildings that want to be accessible. Open-sourcing the positioning engine is a credible alternative that makes it a strong portfolio piece for mobile and embedded roles.

Scope warning

One floor, one building, done properly. Surveying takes longer than anyone plans, so do one floor well rather than a building badly. Do not build AR overlays, a map editor or iOS support in the first version. Multi-floor transitions are a stretch goal for a reason.

Learn what you need on Sefism

Get help with this idea

Most teams underestimate the survey and the evaluation, so bring your building plan and ground truth method to the first conversation. Tauseef can help you tighten the scope, choose a stack your team can finish with, and design an evaluation your examiners will respect. If you are new here, create a free account first.

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.

Discuss this idea
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. Sessions are free for approved Sefism members, and a few slots open each week.

Book a free sessionSee all services

Follow along

New writing, resources and project ideas land here first.