Research On Seam Tracking Technology Based On Laser Structured Light Vision Guidance

In order to meet the requirements of high precision and high efficiency in robot welding production,the seam tracking module based on various sensors has become a key component of robot welding system.In this paper,the key technologies involved in the implementation of robot welding seam tracking based on structured light vision guidance are studied.Firstly,the model of structured light vision-guided seam tracking system is established to clarify the detection principle of seam location information;Structured light measurement model and Eye-in-Hand system model are introduced in turn.Light plane equation parameters are obtained by direct calibration.Hand-eye transformation relationship is solved based on the closeness of coordinate chain in Eye-in-Hand system,and then the overall calibration of the system is completed.In order to extract the position information of the weld spot from the weld image,the pre-processing flow of the weld image is designed,and the image is filtered by the efficient median filtering algorithm.Based on the improved extremum method and gray center of gravity method,the laser fringe center can be extracted quickly and accurately.According to the morphological characteristics of laser stripes in various seam images,the basis for determining the seam points is analyzed,and different extraction algorithms are designed to accurately obtain the pixel coordinate information.Based on the final three-dimensional coordinate information of the discrete welding seam point,the two-step method is used to obtain the weld trajectory.Firstly,the welding seam trajectory is expressed in the form of small continuous straight line segments,and the information of each small line segment on the welding seam trajectory is calculated.Secondly,the transition arc is introduced into the joint form with too large angle between adjacent straight line segments in the welding seam trajectory,and the information of each small line segment is updated.The constraints of the optimal speed under various types of joints between continuous small lines on the weld track are derived,which provides a theoretical basis for determining the optimal of each node in the forward-looking control algorithm for seam tracking motion.The experimental platform of robot welding seam tracking system based on structured light vision guidance is built.The six-axis industrial robot is the main actuator of the system,and the system control tasks are divided into:human-computer interaction task,visual inspection task and motion control task.Based on the real-time extension software KRTS(Kithara Real-time Suite)supported by Windows system,the three tasks with different real-time requirements are integrated into a single multi-core industrial PC,and a shared memory area is created to realize data interaction between tasks.This scheme not only reduces the complexity of system structure,but also saves the cost of system hardware.Finally,based on the experimental platform built in this paper,the tracking performance of the system is verified by the docking V-shaped weld.