Recognition of complex surface shape and welding torch pose control method based on rotating arc

Abstract

The invention discloses a rotating arc based method for recognizing a complex curved surface shape and controlling a welding gun pose. The method comprises specific steps as follows: S1, a rotating arc sensor performs continuous sampling on a weld bead with welding, and sampling points are extracted; S2, a weld bead characteristic curve is calculated and processed according to the sampling points, sot that recognition of a curved surface shape can be finished; S3, a weld bead characteristic section curve is calculated according to the characteristic curve, and pose information of a welding gun is determined; S4, a robot running program is automatically generated according to curved surface shape information and the pose information of the welding gun, and is used for adjusting welding operation in real time. A specific calculation model is established according to basic sampling point information during welding, so that the height and the angle of the welding gun during welding are worked out, the robot running program is automatically generated to adjust the welding operation in real time, and important reference basis is provided for recognition of the complex curved surface and accurate control for the pose of the welding gun while the welding quality and the working efficiency are guaranteed.

Description

technical field [0001] The invention relates to welding automatic control technology, in particular to a method for recognizing the shape of a complex curved surface based on a rotating arc and controlling the position and posture of a welding torch. Background technique [0002] As we all know, the welding of complex curved surfaces is a very important research field in the field of welding automatic control technology. There are a large number of complex curved surfaces in the welding manufacturing of automobiles, trains, and ships. At present, to realize the shape recognition of complex surfaces and the control of welding torch pose, robots with vision sensors are mainly used. In the prior art, it is difficult for the camera to obtain obvious surface topography due to the influence of the complex external environment, such as natural light, welding arc, smoke, splash, etc., during the complex surface shape recognition process of the visual sensor. Moreover, the Canny alg...

AI Technical Summary

Problems solved by technology

In the existing technology, it is difficult for the camera to obtain obvious surface topography due to the complex external environment during the recognition process of the complex surface shape by the visual sensor, such as: natural light, welding arc, smoke, spatter, etc.

Moreover, the Canny algorithm, Laplace algorithm, etc. used by the robot system based on visual sensors are difficult to process complex surface images quickly, which makes it inevitable to lag in time for the pose adjustment of the welding torch.

In addition, the total cost of the robot system of the visual sensor is relatively high. Due to the influence of the complex communication protocol, it cannot be connected conveniently and reliably with the mainstream welding equipment system in the industry in data communication, which has certain limitations in the actual production and application process.

[0003] In actual industrial production, robot teaching or offline programming methods are sometimes used for welding complex workpieces with certain regular surface shapes, but there are a large number of uncertain factors in the welding process, such as the machining accuracy of the welded workpiece groove , heat in the welding process, arc force, magnetic deflection, etc. will affect the accuracy of robot teaching or offline programming.

Moreover, manual teaching takes up the working time of the robot, and manual programming also takes a lot of time for complex calculations, programming and debugging, which greatly reduces work efficiency

In addition, robot teaching or off-line programming cannot require the welding robot to have a certain degree of flexibility, which makes it susceptible to the influence of the external environment and a large number of uncertain factors, which makes the welding robot often deviate from the complex surface shape recognition and welding torch pose control. Generally, the quality will decline or even fail, and the normal and smooth progress of industrial production cannot be guaranteed.

Images