Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How to compute the rotation matrix

  1. Feb 24, 2012 #1
    Hai , I have two matrix , let say X=[ A1 A2 A3
    A4 A5 A5
    A7 A8 A9
    A10 A11 A12]
    and X0=[B1 B2 B3
    B4 B5 B6
    B7 B8 B9
    B10 B11 B12]

    Can any one show me how to compute the rotation matrix from X and X0 ?
    Thank You
     
  2. jcsd
  3. Feb 25, 2012 #2

    chiro

    User Avatar
    Science Advisor

    Hey Renoald and welcome to the forums.

    One problem that you have is that a rotation matrix preserves the dimension and is a basis which means that the matrix has to be square (even if some values don't contribute in the form of zero entries).

    I think it would be helpful if you gave the exact dimensions of your matrices for X and X0. If these are vectors in a Euclidean space, then this is a very well understood problem, but if they are matrices then we will need a more general result.

    So lets say X is a matrix with 3 rows and 2 columns, X will be a 3x2 matrix.
     
  4. Feb 25, 2012 #3
    Hai , Thank You for reply!
    The dimension of X and X0 is 4 x 3 ( Matrix form) and This is not a square matrix.
    Let Say X=RX0 , then the rotation matrix is R . How to compute the R ?
    Thank You .......
     
  5. Feb 25, 2012 #4

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    I am afraid you are going to have to explain what you mean by "rotating" one matrix to another.
     
  6. Feb 26, 2012 #5
    Hi Renoald.

    In [itex]\mathbb R^n[/itex] the term "rotation" usually means an orthogonal matrix with the determinant 1 (the determinant of an orthogonal matrix is always 1 or -1).

    So, as far as I understand your question is: given two [itex]4\times 3[/itex] matrices A and B find a rotation (an orthogonal matrix with determinant 1) R such that B=RA. And your matrices have real entries. Am I correct here?

    Note that such R does not always exists: for 2 real matrices A and B one can find an orthogonal matrix R such that B=RA if and only if [itex]A^T A =B^T B[/itex], where [itex]A^T[/itex] is the transpose of A. So I assume your matrices satisfy this condition; if not, you are lucky, because such R does not exists, and you do not have to do anything :)

    Probably the easiest way to find R is to apply Gram-Schmidt orthogonalization to the columns of one of the matrices (say A). If you know what it is, I can tell you what to do; if not, you have to learn it first.
     
  7. Feb 27, 2012 #6
    Hai , thank you for the reply.
    example i give here is what called as coordinate transformation.
    I not sure Gram-Schmidt orthogonalization can used to resolve this problem or not !
    as what i search by google , the solution given is Helmert transformation.
    Any one have idea about this ?
     
  8. Feb 27, 2012 #7
    Renoald,
    if you want to get help, then STATE the problem first.

    Am I correct, that translated to the mathematical language you problem can be stated like that:
    given two 4×3 matrices A and B (with real entries) find a rotation (an orthogonal matrix with determinant 1) R such that B=RA?

    If that is the statement, then the problem CAN be solved using Hilbert-Schmidt orthogonalization.

    If that is not the statement, you should STATE the problem first: in particular, what do you mean by rotation in 4-dimensional space?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: How to compute the rotation matrix
  1. Matrix in Rotation (Replies: 2)

  2. Rotational Matrix (Replies: 5)

  3. Rotation matrix (Replies: 4)

Loading...