Home /Research /Robot intelligent welding programming based on line structure light sensing
PERCEPTION

Robot intelligent welding programming based on line structure light sensing

Yao Xiao, Bo Zhou, Jiahui Xuan

Year
2018
Citations
5

Abstract

The traditional robot automatic welding programming method relies heavily on the accuracy of CAD model for welding workpiece, so its application is limited for small batch, irregular and multiple type welding tasks. In this paper, with the traditional welding programming software, a new three-dimensional sensing and point cloud modeling approach of the weld is constructed by introducing a line structure light sensor, thus the accurate extracting of weld seams is realized. A general intelligent welding programming method and software system is developed by using the framework of the combination of front-end perception, point cloud processing, and back-end off-line programming. Through modular design, a unified interface is designed for the whole process of welding, from welding seam extraction to robot welding intelligent programming, is realized for different types of sensors, robots and scanning methods. The experimental results showed that the proposed robot intelligent welding programming system is effective, universal and extensible.

Keywords

WeldingRobot weldingComputer scienceRobotModular designInterface (matter)SoftwareArtificial intelligenceComputer visionEngineering drawing

Related papers

Browse all PERCEPTION papers