Dual-sensing dual-wavelength integrated welding seam tracking sensor and deviation rectifying method

Abstract

The invention provides a dual-sensing dual-wavelength integrated welding seam tracking sensor and a method. The two principles of structured light and monocular vision are adopted at the same time to detect the position of a welding seam, a structured light sensor can obtain accurate welding seam position information, a vision sensor can obtain the deflection angle of the welding seam, and therefore leading errors can be corrected in real time. Meanwhile, working wavelengths of the two sensors are located in two different areas with weak arc spectrums respectively, correspondingly, a narrow-band optical filter is also in a double-band-pass mode, only one image needs to be obtained through a color camera, laser stripe information and welding seam area image information can be synchronously obtained through RBG component processing, and therefore all problems caused by asynchronous sampling in sequence are thoroughly solved in principle. The sensors of the two principles share one set of lens and camera, so that the system composition can be greatly simplified, the cost is reduced, the sensor size is reduced, application integration is facilitated, and the actual effect is remarkably improved.

Description

technical field [0001] The invention relates to a welding seam tracking sensor. Specifically, it is a high-performance dual-sensing dual-wavelength integrated welding seam tracking sensor and a deviation correction method. Background technique [0002] With the popularization and low price of factory automation equipment, the welding operation has also developed from traditional manual welding to semi-automatic welding and automatic welding. The welding process with the help of robots is also popular in various fields such as the automobile industry, and it has become a necessary means to rationalize the welding process. In order to achieve the goal of accurate position and no deviation in robot welding, good welding quality consistency, high work efficiency, and the goal of adapting to different types of welding processes and structural changes of different workpieces, it is necessary to detect the center of the welding torch, arc or molten pool in real time during the wel...

AI Technical Summary

Problems solved by technology

However, this line structured light detection method also has its own defects and deficiencies: First, there is a certain distance between the detection position of the existing structured light vision sensor and the actual welding position. Or when the track placement surface is uneven, it is easy to cause the leading error of the welding torch

Its disadvantages are that the system composition is complicated, the cost is high, and the space occupied is large, which is not convenient for the integrated application of welding robots.

In particular, there is inevitably a synchronization error between two or more cameras. In the high-speed dynamic welding process and strong vibration, splash, smoke and other dynamic interference conditions, asynchronous sampling will inevitably lead to measurement errors, thus making the The actual tracking effect is greatly reduced

Images