I have the calibration board fixed on the work station, and I have its world coordinates, entered them in URDF, all good. I used OpenCV's pose estimation module to find the board's pose (rvecs and tvecs) from the camera frame as well. This also is OK. However, I want to get the camera pose (from the world frame) and to do that I think I need to use TF. So basically, we can formulate the problem as follows:
*Given:*
**World -> Board (board in world frame)**
**Board -> Camera (board in camera frame)**
*Find:*
**World -> Camera (world in camera frame, and vice versa, camera in the world frame).**
I have the world and board in my URDF, but since I don't have my camera (which is what I am trying to find out anyway), I don't know how I could utilise TF functionalities.
Does anyone have an experience with this setup?
↧
