Visual detection sensing unit

Abstract

The invention relates to a visual detection unit, in particular to a visual sensing detection unit used for acquiring postwelding surface topography quality in the process of robot laser welding in a visual sensing mode, tracking the welding seam route before welding and monitoring variations of a laser welding molten pool in real time. The invention aims to provide a visual detection sensing unit for robot laser welding. The visual detection sensing unit has the capabilities of simultaneously acquiring postwelding surface topography, tracking the welding seam route before welding and monitoring variations of the laser welding molten pool in real time, and has the advantages of simple and reasonable structure, and convenient operation. The technical scheme provided by the invention is that the visual detection sensing unit comprises a shell arranged on a clamping apparatus. The visual detection sensing unit is characterized in that the inside of the shell comprises a light irradiationdevice, an imaging system, an optical lens assembly, a light reflection device and a mechanical adjusting device.

Description

technical field [0001] The invention relates to a visual detection device, in particular to a visual sensing detection device used for visual sensing in the laser welding process of a robot to obtain the surface topography quality after welding, to track the weld seam route before welding and to monitor the changes of the laser welding molten pool in real time. Background technique [0002] At present, there are many kinds of sensing methods applied to welding robots. Among these sensing methods, due to the advantages of large amount of information, non-contact, fast, high precision and large detection range, the visual sensing method is widely used in welding work. It has become a research hotspot in robotic welding sensors. [0003] Among the existing robotic laser welding vision sensors, most of them use active vision and passive vision to solve the pre-weld tracking and post-weld shape detection and online molten pool monitoring respectively. The active visual detection...

AI Technical Summary

Problems solved by technology

Passive vision mainly uses the laser generated during welding to first reflect the strong light without passing through the linear structured light laser, use the light reduction film to weaken the light, then filter the stray light through the filter, and finally reach the camera to obtain the image; due to the two The visual method uses an independent device; it needs to be used separately, the function is single, and the use is more complicated

[0004] In response to the above problems, existing technicians have combined the two sensors of weld seam tracking and post-weld shape detection by adding an electric slider at the end of the robot. Although the two sensors with single functions are connected together, there is still Complicated manipulation, poor usage stability, inaccurate test results, etc.

Images