Ellipse Rotation: Solving with Normal Rotation Matrix

Physgeek64
Messages
245
Reaction score
11
Hello! Okay- This is a relatively simple problem, but for some reason I'm having huge difficulty with it.

So I have the equation of an ellipse, x^2-6sqrt3 * xy + 7y^2 =16, which I have converted into quadratic form to get (13, -3sqrt3, -sqrt3, 7) and I need to rotate it using the normal rotation matrix in two dimensions (cos, -sin, cos, sin)

But I am struggling to actually apply the rotation matrix- Do apply it to the quadratic form of the matrix? And if so how can I extract the equation of the rotated ellipse from this?

Any help would be GREATLY appreciated! Thank you
 
Physics news on Phys.org
Assume the equation of ellipse you have there to be written in the already rotated coordinate system ##(x',y')##, thus
$$
x'^2-6\sqrt{3} x'y' + 7y'^2 =16
$$
To obtain the expression of this same ellipse in the unrotated coordinate system, you have to apply the clockwise rotation matrix to the point ##(x',y')##. By the way the correct rotation matrix is
$$
\left( \begin{array}{cc}
\cos \theta & \sin\theta \\
-\sin\theta & \cos \theta \\
\end{array} \right)
$$
where ##\theta## is positive for counterclockwise rotation.
 
Last edited:
blue_leaf77 said:
Assume the equation of ellipse you have there to be written in the already rotated coordinate system ##(x',y')##, thus
$$
x'^2-6\sqrt{3} x'y' + 7y'^2 =16
$$
To obtain the expression of this same ellipse in the unrotated coordinate system, you have to apply the clockwise rotation matrix to the point ##(x',y')##. By the way the correct rotation matrix is
$$
\left( \begin{array}{cc}
\cos \theta & -\sin\theta \\
\sin\theta & \cos \theta \\
\end{array} \right)
$$
where ##\theta## is positive for counterclockwise rotation.
Okay, so do I simply apply the rotation matrix to the quadratic form?
 
Physgeek64 said:
I have converted into quadratic form to get (13, -3sqrt3, -sqrt3, 7)
First of all, how did you get that quadratic form, that doesn't seem to lead to the original ellipse equation.
 
blue_leaf77 said:
First of all, how did you get that quadratic form, that doesn't seem to lead to the original ellipse equation.
if you multiple by (x,y)T A (x,y) where A is the matrix stated above. Sorry I don't know how to use matrix notation on here- but the T represents transpose
 
Physgeek64 said:
if you multiple by (x,y)T A (x,y) where A is the matrix stated above.
Well yeah I know that, but if you really carry out the matrix operation
$$
\left( \begin{array}{cc}
x & y \\
\end{array} \right)
%
\left( \begin{array}{cc}
13 & -3\sqrt{3} \\
-\sqrt{3} &7 \\
\end{array} \right)
%
\left( \begin{array}{c}
x \\
y
\end{array} \right) =
\left( \begin{array}{cc}
x & y \\
\end{array} \right)
%
\left( \begin{array}{c}
13x-3\sqrt{3}y \\
-\sqrt{3}x + 7y \\
\end{array} \right) =
13x^2 - 4\sqrt{3}xy + 7y^2
$$
It cannot be the same as the ellipse equation you have in your first post for whatever constant value in the right hand side.
 
blue_leaf77 said:
Well yeah I know that, but if you really carry out the matrix operation
$$
\left( \begin{array}{cc}
x & y \\
\end{array} \right)
%
\left( \begin{array}{cc}
13 & -3\sqrt{3} \\
-\sqrt{3} &7 \\
\end{array} \right)
%
\left( \begin{array}{c}
x \\
y
\end{array} \right) =
\left( \begin{array}{cc}
x & y \\
\end{array} \right)
%
\left( \begin{array}{c}
13x-3\sqrt{3}y \\
-\sqrt{3}x + 7y \\
\end{array} \right) =
13x^2 - 4\sqrt{3}xy + 7y^2
$$
It cannot be the same as the ellipse equation you have in your first post for whatever constant value in the right hand side.
Sorry- that was a typo on my part- both of the non-leading diagonal entries should be -3sqrt3
 
What about the coefficient of ##x^2##?
 
blue_leaf77 said:
What about the coefficient of ##x^2##?

Also a typo- sorry
it is meant to be 13x^2
 
  • #10
blue_leaf77 said:
What about the coefficient of ##x^2##?
I unfortunately don't know how to edit my original post to correct these though
 
  • #11
Physgeek64 said:
Okay, so do I simply apply the rotation matrix to the quadratic form?
What are you asked to do? Are you asked to express the rotated ellipse in quadratic form or in algebraic form?
 
  • #12
blue_leaf77 said:
What are you asked to do? Are you asked to express the rotated ellipse in quadratic form or in algebraic form?
algebraic, but in the unrotated co-ordinate system
 
  • #13
Physgeek64 said:
algebraic, but in the unrotated co-ordinate system
Then you don't need to take a detour by going to the quadratic form since you already have the algebraic form of the original ellipse. So as I said earlier, imagine you have the ellipse in the already rotated coordinate system ##(x',y')##, which means
$$
13x'^2-6\sqrt{3} x'y' + 7y'^2 =16
$$
At the same time, you also have the matrix relation between ##(x',y')^T## and ##(x,y)^T##, related by coordinate rotation. From this, you should be able to obtain expressions for ##x'## and ##y'## in terms of ##x## and ##y## and substitute to the ellipse equation.
 
Back
Top