2

I am currently working on an AR app that scans beacons when there is not enough features. I have 3 beacons and I want to use 2D trilateration to determine the user's location . The position of the 3 beacons is known, as well as the distance between the user and each specific beacon. I have attached a diagram to show. Does anyone know how to code the trilateration algorithm to find the user's location ( vector 2 )

Diagram Pls See to Understand Better

  • Position of (X1,Y1) , (X2,Y2) , (X3,Y3) , D1, d2 and D3 are known (refer to pic^) *
Jonathan Leffler
  • 730,956
  • 141
  • 904
  • 1,278
Yugo Saito
  • 33
  • 3
  • Bluetooth RSSI is very variable, subject to attenuation by obstacles, reflections, changes in air humidity, beacon signal strength, .... In an indoor setting it isn't going to give you a precise position, maybe one side of a room vs another but probably no better and it will wander about. Instead of trilateration consider a simple ML model like KNN which can handle the non-linear variability better. – Ian Mercer Oct 30 '20 at 05:15
  • hi thanks for the suggestion ! But how will using KNN get the user's location? and can i use KNN with beacons? – Yugo Saito Oct 30 '20 at 09:22
  • 1. Setup beacons. 2. Take receiver to each room (or area within a room), record signal strengths. 3. Repeat for multiple positions, angles and humidity values. 4. Match reading to recordings using KNN to identify room. Or if your question is "how do I get a sub-meter accurate position doing this" the answer is "you can't". – Ian Mercer Oct 30 '20 at 22:59
  • hi i came across this research paper https://www.researchgate.net/publication/339664856_An_Improved_Weighted_KNN_Algorithm_for_Indoor_Positioning_using_Beacon_Devices ; it says we are able to use KNN algorithm to determine the user's position . Thanks for your instruction btw :) – Yugo Saito Nov 02 '20 at 07:42

0 Answers0