visp extrinsic calibration (visp_hand2eye_calibration)

asked 2020-03-11 04:58:14 -0500

marko1990 gravatar image

Hello all,

I am trying to make an extrinstic calibration of the camera mounted on the robot tool over the end effector. I have made teh intristic calibration for this purpose already. I am stuck with the extrinsic calibration. From my understanding the result of the extrinsic calibration would be the relation/transform between the camera frame and the end effector frame, which is later needed to move the camera in certain position in which you would want the end effector to do something e.g. to grasp an object. What I do not understand visp_hand2eye_calibration is:

Example An example client is shipped with this package. It feeds different transformations (camera->object and world->hand) to the calibrator which computes the relative transformation between the the camera and the hand.

world->hand transform is more or less clear, it should be a transform between robot base frame and the robot end effector frame. camera->object - transform I do not understand, isn't the whole idea of taking the images of the known object to estimate the camera frame transform and together with the poses of the robot in each image later calculate the relation between the camera frame and the end effector frame? where is the camera->object coming from?

edit retag flag offensive close merge delete