Matrix Transformations around z axis

In summary, a matrix transformation around the z axis is a rotation operation that changes the position and orientation of an object in three-dimensional space. It is different from other transformations as it only affects rotation, while others can involve translation, scaling, and shearing. The purpose of this transformation is to change the orientation of an object, and it is represented mathematically by a 3x3 rotation matrix. However, there are limitations to this transformation, such as only being applicable to three-dimensional objects and the potential for Gimbal lock.
  • #1
concon
65
0

Homework Statement


The matrix A transforms space by rotating counterclockwise around z axis by ∏/4.

-What is A?
-Find a unit vector s.t. Au=u
(make sure the first nonzero is positive)
-for what λ's does Au = λu? write answer as a set {a,b,...}


Homework Equations



Unit vector u = x/norm(x)




The Attempt at a Solution



So I already found A :

[ sqrt2/2 -(sqrt2/2) 0
sqrt2/2 sqrt2/2 0
0 0 1]

which is correct.

How do I find u?

I think once I find u do I follow the process associated with finding eigenvalues
by using det(A-λI) to find λ's?
 
Physics news on Phys.org
  • #2
concon said:

Homework Statement


The matrix A transforms space by rotating counterclockwise around z axis by ∏/4.

-What is A?
-Find a unit vector s.t. Au=u
(make sure the first nonzero is positive)
So I already found A :

[ sqrt2/2 -(sqrt2/2) 0
sqrt2/2 sqrt2/2 0
0 0 1]

which is correct.

How do I find u?
Consider the equation ##Au = u##. This means that ____ is an eigenvalue of ##A## and ____ is a corresponding eigenvector. (Fill in the blanks.)
 
  • #3
jbunniii said:
Consider the equation ##Au = u##. This means that ____ is an eigenvalue of ##A## and ____ is a corresponding eigenvector. (Fill in the blanks.)

This seems like it might be wrong, but I did the usual procedure for finding eigenvalues and got
λ = sqrt(1/2) + sqrt(2)/2
Is this right? I think I might have solved the equation wrong?
 
  • #4
concon said:
This seems like it might be wrong, but I did the usual procedure for finding eigenvalues and got
λ = sqrt(1/2) + sqrt(2)/2
Is this right? I think I might have solved the equation wrong?
That's one of the eigenvalues, assuming you meant ##\sqrt{1/2} + i \sqrt{1/2}##. (It's a complex number.) But there are two others.

The equation ##Au = u## tells you what one of the eigenvalues and eigenvectors must be.
 
Last edited:
  • #5
By the way, there's an easy geometric interpretation that should be helpful: what vectors are unaffected by rotation around the z axis?
 
  • #6
jbunniii said:
By the way, there's an easy geometric interpretation that should be helpful: what vectors are unaffected by rotation around the z axis?

Just thinking, but does that mean that 1 must be one of the eigenvalues?

We haven't been over this in class so I'm just guessing, but is a zero vector unaffected by rotation?
 
  • #7
concon said:
Just thinking, but does that mean that 1 must be one of the eigenvalues?

We haven't been over this in class so I'm just guessing, but is a zero vector unaffected by rotation?
Yes, a zero vector is unaffected by rotation, but a zero vector is also unaffected by any linear transformation (matrix multiplication), so it doesn't give us much information. For this reason, eigenvectors are defined to be nonzero vectors satisfying ##Av = \lambda v## for some ##\lambda##.

What other vectors are unaffected by rotation around the z axis?
 
  • #8
jbunniii said:
Yes, a zero vector is unaffected by rotation, but a zero vector is also unaffected by any linear transformation (matrix multiplication), so it doesn't give us much information. For this reason, eigenvectors are defined to be nonzero vectors satisfying ##Av = \lambda v## for some ##\lambda##.

What other vectors are unaffected by rotation around the z axis?

Wouldn't the vectors e1,e2,e3= {(1,0,0),(0,1,0),(0,0,1)} be unaffected?
 
  • #9
concon said:
Wouldn't the vectors e1,e2,e3= {(1,0,0),(0,1,0),(0,0,1)} be unaffected?
Well, ##e_1## points in the direction of the positive x axis. Is the x-axis affected by rotation around the z axis?
 
  • #10
jbunniii said:
Well, ##e_1## points in the direction of the positive x axis. Is the x-axis affected by rotation around the z axis?
I think that yes it would have to be affected now that I think about it.So then should u = (0,0,1) since e3 points in direction of z?
 
  • #11
concon said:
So then should u = (0,0,1) since e3 points in direction of z?
Exactly.

In three dimensional space, there is always one and only one axis that isn't affected by a non-trivial (i.e., non-identity) rotation. That's the rotation axis.
 
  • #12
OK, that gives you one eigenvalue (1) and corresponding eigenvector ([0, 0, 1]). And the eigenvector is already scaled so it is a unit vector, as requested in the problem statement. So that takes care of the second question, which was:
Find a unit vector s.t. Au=u
(make sure the first nonzero is positive)
Next step is to find the other two eigenvalues:
for what λ's does Au = λu? write answer as a set {a,b,...}
 
  • #13
for what λ's does Au = λu? write answer as a set {a,b,...}
Actually, that question is worded rather poorly. It could be interpreted as referring to the ##u## from the previous part, in which case of course only ##\lambda = 1## is correct and the question seems pointless.

But I interpret it to mean "for what ##\lambda##'s is there a nonzero ##u## such that ##Au = \lambda u##?" In other words, find all of the eigenvalues. Probably best to check with your instructor just to make sure.
 
  • #14
jbunniii said:
Actually, that question is worded rather poorly. It could be interpreted as referring to the ##u## from the previous part, in which case of course only ##\lambda = 1## is correct and the question seems pointless.

But I interpret it to mean "for what ##\lambda##'s is there a nonzero ##u## such that ##Au = \lambda u##?" In other words, find all of the eigenvalues. Probably best to check with your instructor just to make sure.

yes that is what the professor meant. So we already have 1 as eigenvalue. The question specificies that only real values of λ to be included. Earlier in the thread I solved for the eigenvalues and got
λ= sqrt(1/2) + isqrt(1/2) which isn't a real number right?
So should answer just be{1}?
 
  • #15
concon said:
yes that is what the professor meant. So we already have 1 as eigenvalue. The question specificies that only real values of λ to be included. Earlier in the thread I solved for the eigenvalues and got
λ= sqrt(1/2) + isqrt(1/2) which isn't a real number right?
So should answer just be{1}?
Yes, if only real numbers are allowed then the only eigenvalue is 1.
 

1. What is a matrix transformation around the z axis?

A matrix transformation around the z axis, also known as a rotation around the z axis, is a mathematical operation that changes the position and orientation of an object in three-dimensional space. It involves multiplying the coordinates of the object by a specific rotation matrix to achieve the desired transformation.

2. How is a matrix transformation around the z axis different from other transformations?

A matrix transformation around the z axis is different from other transformations because it only affects the rotation of an object in three-dimensional space, while other transformations can also involve translation, scaling, and shearing.

3. What is the purpose of a matrix transformation around the z axis?

The purpose of a matrix transformation around the z axis is to change the orientation of an object in three-dimensional space. This can be useful in computer graphics, robotics, and other fields where precise positioning and orientation is required.

4. How is a matrix transformation around the z axis represented mathematically?

A matrix transformation around the z axis is represented by a 3x3 rotation matrix, which contains the cosines and sines of the desired rotation angle. The rotation matrix is multiplied by the coordinates of the object to achieve the transformation.

5. Are there any limitations to matrix transformations around the z axis?

Yes, there are limitations to matrix transformations around the z axis. For example, they can only be applied to objects in three-dimensional space and cannot change the shape or size of an object. Additionally, repeated rotations around the z axis can lead to Gimbal lock, which can cause unexpected results.

Similar threads

  • Calculus and Beyond Homework Help
Replies
1
Views
996
  • Calculus and Beyond Homework Help
Replies
4
Views
1K
  • Calculus and Beyond Homework Help
Replies
4
Views
839
  • Calculus and Beyond Homework Help
Replies
5
Views
1K
  • Calculus and Beyond Homework Help
Replies
1
Views
608
  • Calculus and Beyond Homework Help
Replies
1
Views
1K
  • Calculus and Beyond Homework Help
Replies
2
Views
512
  • Calculus and Beyond Homework Help
Replies
11
Views
1K
  • Calculus and Beyond Homework Help
Replies
10
Views
2K
  • Calculus and Beyond Homework Help
Replies
10
Views
1K
Back
Top