Hand-Eye Calibration Concept and Theory
This articles covers the concept and theory of hand-eye calibration without going into the details of the Zivid hand-eye calibration software.
Note
If you are looking for a practical guide on how to run or integrate hand-eye calibration with Zivid software, skip to How to select Zivid Calibration Object.
Where to Start?
Hand-Eye System Configurations
Explains the two standard camera placements in vision-guided robotics: eye-in-hand and eye-to-hand.
-
Breaks down the hand-eye calibration task by first explaining how a robot needs object poses relative to its base and how a camera’s coordinate system ties into that. It covers both eye-to-hand and eye-in-hand setups, showing the required coordinate transforms.
-
Describes the mathematical solution to the hand-eye calibration problem, and how the required coordinate transforms can be estimated from the data collected during the calibration process.
-
Discusses typical calibration objects used for hand-eye calibration and their geometry.
-
Describes the typical workflow of performing hand-eye calibration, including data collection and software processing.
Hand-Eye Calibration Residuals
Explains the concept of hand-eye calibration residuals and how they can be calculated to evaluate the performance of the calibration.
To get into the hand-eye calibration theory, start with the Hand-Eye System Configurations.