How to calculate 2D Trilateration Step by Step

  • Context: Graduate 
  • Thread starter Thread starter Jose F Gomez D
  • Start date Start date
  • Tags Tags
    2d Triangulation
Click For Summary

Discussion Overview

The discussion revolves around the mathematical and computational aspects of 2D trilateration, specifically how to determine the coordinates of an unknown point based on distances from three known reference points. Participants explore different methods, calculations, and the implications of the triangle inequality on the problem.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant describes their approach to coding trilateration, detailing the calculations for unit vectors and distances, but expresses confusion about the next steps.
  • Another participant suggests a straightforward method using squared distance equations to derive the coordinates of the unknown point, proposing to combine equations to solve for x and y.
  • A different participant challenges the validity of the proposed method, stating that the calculations do not yield correct results for the distances provided.
  • Some participants point out that the distances given violate the triangle inequality, indicating that no such point can exist based on the specified distances from the reference points.
  • One participant suggests using a graphical method to visualize the problem, indicating that the intersection of circles representing distances may reveal the solution or lack thereof.
  • A participant shares a link to a paper outlining a trilateration algorithm, indicating alternative approaches to the problem.

Areas of Agreement / Disagreement

There is disagreement among participants regarding the validity of the proposed methods and calculations. Some participants assert that the distances provided do not allow for a solution, while others attempt to derive coordinates using different mathematical approaches. The discussion remains unresolved with competing views on the feasibility of finding a solution.

Contextual Notes

Participants note that the distances from the reference points may not satisfy the triangle inequality, which is critical for the existence of a solution in trilateration. This introduces uncertainty regarding the applicability of the methods discussed.

Jose F Gomez D
Messages
6
Reaction score
0
I am trying to code Trilateration but I am not familiar with the math behind it. even though there is an explanation on Wikipedia in this link https://en.wikipedia.org/wiki/Trilateration
and a almost better explanation in here https://stackoverflow.com/questions/9747227/2d-trilateration?answertab=active#tab-top I am still confused on how to do it

I have P1, P2 and P3 as my reference points, and X as the point to find
and I want the distance from P1 to X = 8, P2 to X = 10 and P3 to X = 11

P1 (10,10)

P2 (24,24)

P3 (23,19)

so I have to calculate the unit vector in the direction from P1 to P2, following the formula ex = (P2 - P1) / ‖P2 - P1‖ meaning

ex,x = (P2x - P1x) / sqrt((P2x - P1x)^2 + (P2y - P1y)^2)

ex,y = (P2y - P1y) / sqrt((P2x - P1x)^2 + (P2y - P1y)^2)

ex,x = (24 - 10) / sqrt((24 - 10)^2 + (24 - 10)^2) = 14/sqrt(14^2+14^2) = 14/sqrt(392) = 14/20 = 0.7

ex,y = (24 - 10) / sqrt((24 - 10)^2 + (24 - 10)^2) = 14/sqrt(14^2+14^2) = 14/sqrt(392) = 14/20 = 0.7

then calculating the signed magnitude of the x component of the vector from P1 to P3 by i = ex(P3 - P1)

ix = 0.7*(23-10) = 0.7*13 = 9.1

iy = 0.7*(19-10) = 0.7*9 = 6.3

now calculating the the unit vector in the y direction from the formula
ey = (P3 - P1 - i*ex) / ‖P3 - P1 - i*ex‖**but here how do I continue?**I am assuming to continue like this

ey,y = (P3y - P1y - iy*ex,y) / sqrt((P3x - P1x - ix*ex,x)^2+(P3y - P1y - iy*ex,y)^2)

ey,x = (P3x - P1x - ix*ex,x) / sqrt((P3x - P1x - ix*ex,x)^2+(P3y - P1y - iy*ex,y)^2)

ey,y = (19 - 10 - 6.3*0.7) / sqrt((23 - 10 - 9.1*0.7)^2+(19 - 10 - 6.3*0.7)^2)

ey,x = (23 - 10 - 9.1*0.7) / sqrt((23 - 10 - 9.1*0.7)^2+(19 - 10 - 6.3*0.7)^2)

ey,y = 0.6

ey,x = 0.8

then the distance between the centers P1 and P2 is ‖P2 - P1‖

d = ‖P2 - P1‖ = sqrt((P2x - P1x)^2 + (P2y - P1y)^2) = 20

then the signed magnitude of the y component from j = ey(P3 - P1)

jxx = 0.6*(23-10) = 0.6*13 = 7.8

jxy = 0.8*(23-10) = 0.8*13 = 10.4jyx = 0.6*(19-10) = 0.6*9 = 5.4

jyy = 0.8*(19-10) = 0.8*9 = 7.2

then finally from https://stackoverflow.com/questions/9747227/2d-trilateration?answertab=active#tab-top

x = (r12 - r22 + d2) / 2d

y = (r12 - r32 + i2 + j2) / 2j - ix / jx = (8^2 - 10^2 + d^2) / 2*d = (8^2 - 10^2 + 20^2) / (2*20) = 9.1

y = (8^2 - 11^2 + i^2 + j^2) / (2*j - (ix / j))

**What do I do?**

please help
 
Mathematics news on Phys.org
I have no idea what this method involves. However there is a straightforward approach as follows, with X=(x,y)
1)\ (x-10)^2+(y-10)^2=64 (distance to P1 squared)
2)\ (x-24)^2+(y-24)^2=100 (distance to P2 squared)
3)\ (x-23)^2+(y-19)^2=121 (distance to P3 squared)
Combine equations: 1) minus 3) to get 4), 2) minus 3) to get 5). 4) and 5) are linear in x and y. Solve and check to see if they satisfy 1), 2), and 3).
 
nop it does not work in this way

P1 = x^2-20x+100+y^2-20y+100=64

P1 = x^2-20x+y^2-20y=-136P2 = x^2-48x+576+y^2-48y+576=100

P2 = x^2-48x+y^2-48y=-1052P3 = x^2-46x+529+y^2-38y+361=121

P3 = x^2-46x+y^2-38y=-769P4=P1-P3(x^2-20x+y^2-20y=-136)-( x^2-46x+y^2-38y=-769)

(x^2-20x+y^2-20y)-( x^2-46x+y^2-38y)=633

P4 = (26x+18y)=633P5=P2-P3(x^2-48x+y^2-48y=-1052)-( x^2-46x+y^2-38y=-769)

(x^2-48x+y^2-48y)-( x^2-46x+y^2-38y)=-283

(-2x-10y)=-283

P5 = (2x+10y)=283Substitution

(26x+18y)=633 y=(633-26x)/18

(2x+10y)=283 (2x+10((633-26x)/18))=283

(2x+((6330-260x)/18))=283 36x+6330-260x=283*18

6330-224x=5094 1236=224x

x= 5.52
y=(633-26(5.52))/18= 27.2
replacing x and y on P1,2,3

x^2-20x+y^2-20y=-136
〖5.52〗^2-20*5.52+〖27.2〗^2-20*27.2=315.91
wrong

x^2-48x+y^2-48y=-1052
〖5.52〗^2-48*5.52+〖27.2〗^2-48*27.2=351.75
wrong

x^2-46x+y^2-38y=-769
〖5.52〗^2-46*5.52+〖27.2〗^2-38*27.2=372.79
wrong
 
Jose F Gomez D said:
(x^2-20x+y^2-20y=-136)-( x^2-46x+y^2-38y=-769)
That is a really strange way of writing it.

The approach is right, if your result is wrong you did a mistake somewhere in between.
 
No, the calculation is right, x and y works with p4 and p5 but not with p1,2,3, it will never work I have checked online, this does not count, at the end I need trilateration not this
 
I checked the equations, your initial equations don't have a solution to start with.
Here are p1 and p2. The problem is easy to track down:
Jose F Gomez D said:
and I want the distance from P1 to X = 8, P2 to X = 10 and P3 to X = 11

P1 (10,10)

P2 (24,24)
P1 and P2 have a distance of 14*sqrt(2) > 18. There is no such point.
 
OMG, I do not have ecuations, I have coordinates in an x y Cartesian map, p1 is located in 10,10
 
Yes, and P2 is located at (24,24), that is 19.8 away from P1. There is no point that is 8 away from P1 and 10 away from P2, that would violate the triangle inequality.
 
Using graph paper and a compass, one can easily see if there is a solution. Place each point on the paper and use it a center of a circle with radius equal to the distance to the unknown point. If the three circles have a common point of intersection that is the solution. Otherwise (as in this case) there is none.
 
  • #10

Similar threads

  • · Replies 24 ·
Replies
24
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
1
Views
2K
Replies
2
Views
6K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
4K
Replies
3
Views
2K
  • · Replies 41 ·
2
Replies
41
Views
6K