I am trying to understand the implementations and the relative performance of various classical and new Hand-eye calibration approaches.
Note: The following discussion assumes implementations are in Matlab for the AX=XB hand-eye calibration approach.
I have been able to successfully implement Wengerts and Lazax implementation of Tsia approach. These open-source implementations take in the gripper and the camera poses and further performs the specific operations. However, I am facing trouble understanding Milli Shah's implementation of Tsai and Liangs method. The implementation takes in AA (for gripper) and BB (for the camera), which are definitely not the relative (i to j) gripper and camera poses but some processed version of these.
Can anybody please explain this AA and BB in more detail. In case somebody has an extraction code for this, it would be helpful Thanks