Hand-eye calibration optimization method and system adaptive to robot welding scene

By combining physical touch and visual axis alignment with adaptive image acquisition and collision detection, the complexity and safety issues of hand-eye calibration in robotic welding scenarios are solved, enabling efficient and safe conversion of calibration data into robot motion.

CN122274532AActive Publication Date: 2026-06-26DAHE ZHONGBANG (XIAMEN) INTELLIGENT TECH CO LTD +1
5 Cites 0 Cited by

Patent Information

Application Number
CN202610603425.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-06
Publication Date
2026-06-26
Estimated Expiration
2046-05-06

AI Technical Summary

Technical Problem

Existing hand-eye calibration methods in robotic welding scenarios suffer from problems such as difficulty in clearly capturing weld seam trajectories, the risk of collision between the welding torch and the workpiece, and the need for secondary transformation of the transformation matrix, resulting in complex and inefficient application of calibration results.

Method used

By employing a physical touch and vision axis alignment method, the robot's end effector touches the center of the calibration block, and combined with the vision camera's posture adjustment, the camera tool coordinate system parameters are directly calculated. This simplifies the calibration process and introduces an adaptive image acquisition and collision detection mechanism to ensure the accuracy and security of the calibration data.

Benefits of technology

It achieves high-precision and safe conversion of calibration data into robot motion, reduces the difficulty of operation, avoids the risk of welding torch collision, and the calibration results can be directly applied to the robot controller without the need for complex coordinate transformation.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a hand-eye calibration optimization method and system adaptive to a robot welding scene, and relates to the field of robot vision calibration, hand-eye calibration is decomposed into independent position measurement and attitude measurement, and the method comprises the following steps: firstly, physically touching the center of a calibration block by using an end effector to obtain a reference point position; then, based on a point cloud processing technology, actively adjusting the posture of the camera to enable the Z axis to be collinear with the normal of the calibration block and the X axis to be collinear with the feature direction of the calibration block, and aligning the point cloud center with the center of the calibration block under the standard posture to obtain a working point posture; and finally, displacement and attitude parameters are calculated, and tool coordinate system parameters capable of being used by a robot controller are directly output. According to the method, physical interaction is used for replacing vision to solve translation, active posture adjustment is used for replacing numerical optimization to solve rotation, the technical threshold and application complexity of hand-eye calibration in the welding scene are remarkably reduced, complex matrix secondary conversion is avoided, and the intuition and reliability of the calibration process are improved.
Need to check novelty before this filing date? Find Prior Art