engasa/indoornav

A research project in indoor navigation using IBeacon technology, with a focus on the blind or partially sighted

Features

• Android application

  • Indoor location tracking using iBeacon
    • Proximity based tracking based on single iBeacon proxmity
    • Full 2d tracking based on triangulation of >= 3 iBeacons
  • Sight impaired navigation (Haptic or Audio based)
  • Sight impaired friendly UI navigation
    • Menu to select a location, (IE. Lakeview Market, Orchard Park Mall)
    • Menu to select a section in said location, (IE. Product section, Food Court)

• Server side application to define locations and upload new locations

  • Hosts locations for android application to download
  • Ability to upload/define new loctations

Requirements

Non-functional Requirements

• User should be able to navigate app without sight
• User should be able to select between haptic and audio feedback
• User should be able to select their location
• User should be able to select a section in their location to navigate to
• User should be able to get feedback on how far they are from their destination
• User should be able to cancel navigation and re-select location or section
• A server should host config details for locations so that users can easily navigate a new location

functional requirements

• App should be able to know its position within +-2 meters of the actual position indoors
• App should be easily deployable to a new location through a config file
• Config file documented be clearly defined and documented

Challenges

• iBeacon distance accurracy
• iBeacons have an accuracy of +- 1 Meter from within 10 meters, and worse if the device is > 10 meters away. This will present a challenge in getting accurate location data while using both triangulation and basic proximity based tracking.
• To combat this an algorithm that averages the readings over time and then triangulates could be use, this is going to be one of the primary areas of research for the project
• Determining a viable method of providing feedback to users who are sight impaired
• Audio feedback is an easy solution, but not preferred as it is disruptive
• A better solution would be to develop a haptic feedback system where different vibration patterns mean different things (Left or Right)

Todo (In order or importance)

1. Basic android app that interfaces with the IBeacon
2. Add menus of locations to app and develop UI
3. Heading and directional tracking (IE, N-S-E-W) in order to give turn based Navigation
4. Navigational feedback for a user from the app (Haptic or voice)
5. Basic location tracking using IBeacon (Simple proximity based tracking)
6. Improve direction
7. Location triangulation using IBeacon (More exact proximity based tracking)
8. Pinpoint navigation of sections in a store (Lakeview market test)
9. Server/client communication between mobile device and server for configuration data and additional tracking information
10. Interpret location configuration files from server on device
11. Scroll map with finger

Timeline

Todo Item	Scheduled Completion	Status
Basic App / Basic Diagrams / Features / Requirments	September 29th, 2015	September 29th, 2015
Menus and UI	October 13th, 2015	N/A
Heading/direction tracking	October 20th, 2015	Completed
Testing/development of feedback system	November 3rd, 2015	Completed
Implementation of feedback system into application	November 24th, 2015	Incomplete Completed
Triangulation testing and algorithm development	December 8th, 2015	N/A
Implementation of the triangulation method into application	December 15th, 2015	N/A
Prototype with ability to navigate from point to point	Jan 5th, 2015	N/A
Create a configuration format for locations and interpret on device	Jan 12th, 2015	N/A
Setup up location configuration server for device to pull locations from	Jan 17th, 2016	N/A
Extra Time (Possibly increase test coverage/fix bugs)	Feb 2nd, 2016	N/A
Thesis	March 15th, 2016	N/A
Presentation Prep	April 5th, 2016	N/A
Presentation	April 14th, 2016	N/A

Distractions

• Christmas Break, 1-2 weeks off
• Reading Break second semester, 1 week off, feb 15-19th 2016

UML

Class Diagram

Sequence Diagram for app launch

Weeks from january 26th to March 1st

• Week 1: Positioning with ibeacons
• Week 2: Realistic map of real store, improving direction estimates using ibeacon with heading sensor
• Week 3: routing to user selected area (with some search), show notifications near IBeacons
• Week 4: Blind, ui improvements
• Week 5: Sensor fusion