MathGroup Archive 2008

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: how to determine the center or foci of an ellipse from a slope

  • To: mathgroup at smc.vnet.net
  • Subject: [mg88979] Re: how to determine the center or foci of an ellipse from a slope
  • From: dh <dh at metrohm.ch>
  • Date: Thu, 22 May 2008 06:17:09 -0400 (EDT)
  • References: <g11r0s$a87$1@smc.vnet.net>


Hi John, Thomas,???

Having a "slope field" is a differential equation. Specifying an initial 

condition, we are able to integrate it and get one of the many possible 

ellipses. Having an ellipse, you can determine the center.

First we must create a function with input: x,y and output: slope, call 

it slope[{x,y}]. If your data is measured on a rectangular grid, the 

function "Interpolation" will do this.

Next we specify the DE:

eq={pos'[t]==slope[pos[t]], pos[0]==pos0}

where pos[t] is the position (a 2 vector) and pos0={x0,y0} are suitable 

initial conditions.

Now we can solve:

NDSolve[eq,p[t],{t,0,tmax}]

where tmax must be chosen according to your data.

Last we must determine the center. One way to do this is to remember 

that an ellipse is a quadratic curve that can be described e.g. by:

a x^2+2 b x y+ c y^2 + 2 d x + 2 f y + g == 0

the parameters a,b,..g can be fitted using some points {x,y} of your 

previous solution. With these parameters, we get the center:

x0=(cd-bf)/(b^2-ac)

y0=(af-bd)/(b^2-ac)

(you may look this up e.g.: http://mathworld.wolfram.com/Ellipse.html)

hope this helps, Daniel



John Kellerham wrote:

> Hi guys,

> I have a slope field (2d-array), which is defined

> by ellipses (can be arbitrarily tilted), which all have the 

> same foci and center but different major and minor half axes. 

> In fact, I obtained the tangent at each point by measurements.

> 

> Does somebody have an idea on how to approach the problem 

> of finding the center or the foci of the ellipses? I think 

> there is no exact solution to the problem?

> 

> Best Regards,

> Thomas

> 





-- 



Daniel Huber

Metrohm Ltd.

Oberdorfstr. 68

CH-9100 Herisau

Tel. +41 71 353 8585, Fax +41 71 353 8907

E-Mail:<mailto:dh at metrohm.com>

Internet:<http://www.metrohm.com>




  • Prev by Date: Re: Interpolation and plot doing strange things with mathematica
  • Next by Date: Re: Re: How to connect one object to the other?
  • Previous by thread: Re: how to determine the center or foci of an ellipse from a slope
  • Next by thread: change a system of PDEs to ODEs via traveling wave