1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Euler Angle transformation, help!

  1. Jul 3, 2013 #1
    I'm doing a research project currently and basically what I have is a camera measuring a probe. I have designed the camera to give the orientation of the probe using euler angles in the camera's frame of reference. This was working for most of my data, but now I need a 3-D visualization of what the camera on the tip of the probe would see. In order to do that, I now need to know the euler angles in the probe's frame of reference.

    I know how to make rotation transformations using euler angles for vectors, but for some reason this problem is confusing me. I need euler angles in one frame where I have the euler angles in another frame. Suggestions on how to make this work are greatly appreciated, thanks!

    EDIT: An easy way would be IF I KNEW the angles between the two coordinate systems, but unfortunately one of the coordinate systems (the probe) is constantly changing. So it is a little more complex than simply adding angles and subtracting them. All I need though I guess, is a formula that gives me the instantaneous transformation. Then I can just put that into my MATLAB code and I'll have all the data I need.


    - Vector between two coordinate systems
    - Orientation of probe at any instance of time (but changing in time)
    - Euler angles of probe in camera's reference frame
    - Rotation matrix between both coordinate systems at any given instance of time
    Last edited: Jul 3, 2013
  2. jcsd
  3. Jul 3, 2013 #2
    Think I figured it out. All I had to do was write a program that constantly gave the difference in the euler angles of the two coordinate systems. Then I took the original vector (pointing in the -z direction of my probe's coordinate system) and did the inverse transformation using the inverse of the rotation matrix given by the euler angles. Finally, I did took that vector and normalized it so I had a unit vector in my new coordinate system pointing in the correct direction.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook