Teaching programming, off-line programming and autonomous programming.
(1) programming teaching
Teaching programming means that the operator uses the teaching board to move the welding torch at the end of the robot to track the weld and record the weld trajectory and welding process parameters in time, and the robot teaches and reproduces the welding process point by point according to the recorded information. This method of recording and reproducing torch posture point by point requires the operator to act as an external sensor, and the robot itself lacks external information perception, so its flexibility is poor. Moreover, for weldments with complex structures, operators need to spend a lot of time teaching, and the programming efficiency is low. When the welding environment parameters change, it is necessary to teach the welding process again, which can not adapt to the changing welding objects and tasks, and the welding accuracy is poor.
(2) Offline programming
Off-line programming adopts partial sensing technology, mainly relying on computer graphics technology, establishes the working model of the robot, simulates the programming results with three-dimensional graphics animation to test the reliability of the programming, and finally transmits the generated code to the robot control cabinet to control the operation of the robot. Compared with teaching programming, off-line programming can reduce the working time of robot and simplify programming with CAD technology. The research on off-line programming technology of robots abroad is mature, and all industrial robot manufacturers are equipped with their own off-line programming software systems. For example, ABB's Robot studio simulation programming software can do both simulation analysis and offline programming. Off-line programming can build a simulated welding environment, and according to the working conditions, use CAD technology to build the corresponding geometric models of fixtures, parts and tools. However, due to the lack of sensing data of real welding environment, the geometric model can only partially describe the real welding target, and the deviation must be adjusted during the welding process. Therefore, off-line programming is difficult to describe the real three-dimensional motion, and it is not particularly reliable. Real-time deviation control must be carried out in the welding process to meet the requirements of welding process.
(3) Independent programming
Autonomous programming technology is the basis of realizing robot intelligence. The application of various external sensors in autonomous programming technology enables the robot to perceive the real welding environment in all directions, identify the welding table information and determine the process parameters.
Autonomous programming technology does not need heavy teaching, which reduces the working time of the robot and the labor time of workers, and does not need to correct the deviation in the welding process in real time according to the information of the workbench, which greatly improves the autonomy and adaptability of the robot and becomes the development trend of the robot in the future.
At present, the commonly used sensors include vision sensor, ultrasonic sensor, arc sensor and contact sensor, which make the robot have vision, hearing and touch.
Robot vision sensor mainly uses CCD (Charge Coupled Device) camera to simulate human eyes to obtain external information, which has the advantages of no contact with workpiece, anti-electromagnetic interference, high detection accuracy and rich information. Ultrasonic sensors are cheap and have good ranging directionality, but ultrasonic waves are easily attenuated by the interference of welding noise and protective airflow, which affects the measurement accuracy. The arc sensor makes full use of the arc parameters in the welding process to measure the weld seam, and the distance between the welding torch and the workpiece can be calculated without additional sensors. It is widely used in the welding of symmetrical groove welds such as V-shaped welds, but it has no good detection ability for complex welds. The contact sensor relies on the probe to move along the weld, and the deviation between the welding gun and the weld can be obtained by detecting the deviation of the probe. The sensor is low in price, simple in principle and easy to realize. However, with the aggravation of probe wear and deformation, the detection accuracy gradually decreases, and the detection ability is average for complex welds and high-speed welding occasions.
Instead, the vision sensor collects natural light welding images, laser structured light images and arc images. Laser sensor has good monochromaticity and high brightness, which plays a good auxiliary role in visual acquisition of welding process and has good detection ability for complex welds. Therefore, the welding robot with visual inspection ability can better adapt to environmental changes and realize robot intelligence.