Skip to content

Latest commit

 

History

History
20 lines (17 loc) · 1.68 KB

readme.md

File metadata and controls

20 lines (17 loc) · 1.68 KB

Operation steps:

EyeInHand:

  1. Import camera parameters from param.txt into the input_EyeInHand folder.
  2. Capture 8 BMP and TIFF images with different poses into the input_EyeInHand folder. Import the positions of the robot arm under 8 different poses (order: Rx, Ry, Rz, X, Y, Z).
  3. By default, use a calibration board with a 20mm center distance (link). Otherwise, modify CircleObjectPointsGenerate function accordingly.
  4. Run the calib EyeInHand() function under calib.py
  5. Check the "output_EyeInHand/result.txt" (order: rotation_matrix[0:9], traslation_vector[0:3])
  6. Result verification: open .xyz files generated in the "output_EyeInHand/" folder into CloudCompare to check the overlap of the centers.

EyeToHand:

  1. Import camera parameters from param.txt into the input_EyeToHand folder.
  2. Capture 5 BMP and TIFF images with different poses into the input_EyeToHand folder. Import the positions of the robot arm under 5 different poses (order: Rx, Ry, Rz, X, Y, Z).
  3. By default, use a calibration board with a 40mm center distance. Otherwise, modify CircleObjectPointsGenerate function accordingly.
  4. Run the calib EyeTonHand() function under calib.py
  5. Check the "output_EyeToHand/result.txt" (order: rotation_matrix[0:9], traslation_vector[0:3])
  6. Result verification: open .xyz files generated in the "output_EyeToHand/" folder into CloudCompare to check the overlap of the centers.