Hand-Eye Calibration

Hand-eye calibration determines the exact spatial transformation between a camera and a robot, so that positions detected in an image can be converted into robot coordinates for picking, placing or inspection.

There are two set-ups. In eye-in-hand, the camera is mounted on the robot and the calibration finds the transform from the robot flange to the camera. In eye-to-hand, the camera is fixed and the calibration finds the transform from the robot base to the camera. The robot moves a calibration target, or moves the camera relative to one, through several poses, and the transform is solved from the recorded robot poses and image measurements, classically formulated as the equation AX = XB.

Accurate hand-eye calibration underpins vision-guided robotics, including bin picking, conveyor picking, depalletising and inspection with robot-mounted cameras. Most vision and robot vendors provide calibration routines and targets.

Results depend on the camera's intrinsic calibration, target quality, the diversity of poses and the robot's accuracy. Any change to the camera mounting, lens focus or robot tool definition invalidates the calibration, so it should be verified after maintenance or collisions, ideally by measuring reference points across the working area.

Key points

Finds the transform between camera and robot coordinates
Eye-in-hand mounts the camera on the robot; eye-to-hand fixes it in the cell
Classically formulated as the AX = XB problem
Must be repeated after camera, lens or tool changes

Where AiVibe comes in

AiVibe designs and manufactures the AiAmbA AI Factory, whose edge devices and AI agents let people talk to robot controllers in plain language. Robotics perception is an AiAmbA use case, and robot safety functions follow ISO 10218 and never depend on the AI layer.

Explore AiVibe’s work in Robotics & Physical AI →

Related terms

Terms that refer to Hand-Eye Calibration

Robot Repeatability and AccuracyRobotics & Physical AI