Study of image processing for V-shape groove and robotic weld seam tracking based on laser vision

Single-stripe laser was applied to acquire V-shape groove contour information. Most of arc light and splash noise was removed and stripe laser image was kept by wavelet transform. Half-threshold algorithm was used for image segmentation and stripe laser image was gotten after refining. Weld seam center position was identified and extracted by extreme curvature corner detection method. The location of torch was detected to accord the frequency of computer program with robot program and serial communication program. The tracking experiments of sidelong, reflex and curve weld line show that the system can meet the demand of the tracking precision under normal welding conditions.

Seam Tracking and Visual Control for Robotic Arc Welding Based on Structured Light Stereovision

A real-time arc welding robot visual control system based on a local network with a multi-level hierarchy is developed in this paper. It consists of an intelligence and human-machine interface level, a motion planning level, a motion control level and a servo control level. The last three levels form a local real-time open robot controller, which realizes motion planning and motion control of a robot. A camera calibration method based on the relative movement of the end-effector connected to a robot is proposed and a method for tracking weld seam based on the structured light stereovision is provided. Combining the parameters of the cameras and laser plane, three groups of position values in Cartesian space are obtained for each feature point in a stripe projected on the weld seam. The accurate three-dimensional position of the edge points in the weld seam can be calculated from the obtained parameters with an information fusion algorithm. By calculating the weld seam parameter from position and image data, the movement parameters of the robot used for tracking can be determined. A swing welding experiment of type V groove weld is successfully conducted, the results of which show that the system has high resolution seam tracking in real-time, and works stably and efficiently.

Seam Tracking Technology for Hyperbaric Underwater Welding

Automatic weld seam tracking technology to be used in hyperbaric underwater damaged pipeline repair welding is much more important, because of poor bevel preparation and severe working condition. A weld seam tracking system based on digital signal processing（DSP） passive light weld image processing technology has been established. A convenient charge coupled device（CCD） camera system was used in the high pressure environment with the help of an aperture and focus altering mechanism to guarantee overall image visibility in the scope of pressure below 0.7 MPa. The system can be used in the hyperbaric environment to pick up the real welding image of both the welding arc and the welding pool. The newly developed DSP technology was adopted to achieve the goal of system real time characteristics. An effective weld groove edge recognition technique including narrow interesting window opening, middle value wave filtering, Sobel operator weld edge detecting and edge searching in a defined narrow area was proposed to remove the guide error and system accuracy was ensured. The results of tracking simulation and real tracking application with arc striking have proved the validity and the accuracy of the mentioned system and the image processing method.

The characteristics of welding currents in a rotating arc leadtandem GMAW process and the weld-seam tracking

Tandem gas metal arc welding （ T-GMAW） process shows a high deposition rate that up to three times o f the single electrode GMAW, so the welding speed could be significantly increased in this process. However, the majority o f this process applications are based on the pre-programmed robotic welding, which does not allow them to track the seam real-time during welding. Rotating arc sensor, sensing the seam position by detecting the changing of welding currents, has been widely adopted in the automatic robot welding process. It is proposed in this paper to integrate the rotating arc sensor with a trailing torch to develop a new approach of rotating arc lead tandem gas metal arc welding （RLT-GMAW） process. The characteristics of the welding currents in the proposed new welding process were firstly studied, and then a self-turning fuzzy control seam tracking strategy was developed for the mobile robot automatic welding. The experimental results showed that the proposed RLT-GMAW process had an excellent seam tracking performance and high welding deposition rate. Even if there were some electromagnetic interactions between the two arcs, the deviation of the welding seam could also be reflected by the fluctuation of the welding currents on the leading arc once the correct welding parameters were selected. Based on the detected deviation, the welding tracking experiments showed that the proposed self-turning fuzzy controller had a good performance for the RLT-GMAW process seam tracking.

Arc sensing system for automatic weld seam tracking(Ⅰ)——Mathematic model

Arc sensor, one of the most popular and effective sensors for automatic seam tracking of arc welding, is systematically studied. A mathematic model of the static and the dynamic properties of the sensor derived on the basis of control theory and experimental method are described. The model extends our understanding of arc sensor from qualitative level to quantitative level. This is important for its design and further development.

Arc sensing system for automatic weld seam tracking (Ⅱ)——Signal processing

Due to violent disturbance of the welding arc, signal processing is the key problem of application of the sensor. By means of the new technique the arc sensor can recognize not only V-groove but also lap joint and butt joint. The sensor has good recognition ability even for welding process with very large current disturbance, e.g. pulsed arc and short circuit welding process, etc. The proposed technique is developed on the basis of modern digital filtering theory and mathematic transformation of the signals from time domain into frequency domain.

Study on contactless ultrasonic sensor and seam tracking

The sensitivity of contactless ultrasonic sensor is improved from 3.3% to 26% through acoustic and electric matching technique in this paper.Along dual crest lines auto seam tracking is firstly realized with contactless ultrasonic sensing.The focused acoustic lens is accurately designed, and a sonic beam with 0.5 mm in diameter is achieved. By means of software and hardware technique the accuracy of seam tracking with ultrasonic sensing is 0.5 mm in lateral direction and 0.2 mm in longitudinal direction respectively.

An adaptive thresholding for weld seam image recognition in real time

Image prooessing of wehl seam in real time is an importunity to make welding rohot be able to track weld seam. The algorithm described in this paper combines some image technologies, such as modified Sobel edge detector and Hough transformation function, and especially the thresholds for image processing are ore aled adaptively by Ineans of a neural network. aests proved that this algorithm has a high reliability and rapidity in distinguishing the position of weld seam even with noises. The algorithm can be used ac the basic program .for robot to track welding seam and furthermore for calculating 3 dimensional information plan robot movement automatically.

基于电流多特征融合的窄间隙P-GMAW摆动电弧传感焊缝跟踪方法

电弧传感是实现窄间隙焊接焊缝跟踪的主要方式之一。针对电弧传感稳定性差、可靠性低的问题,提出融合摆动周期内电流信息的多个统计学特征进行偏差提取的方法,克服单一数据特征容易受到电弧稳定性影响、导致传感精度下降的问题。首先,提取电流信号的多个时域特征,计算特征矩阵与偏差矢量相关性;然后,取相关性高的特征用主成分分析的方法进行融合,取前两个主成分作为观测数据;最后,基于多分类的支持向量机模型对其进行分类试验。试验结果表明最大误差为0.2 mm,0.1 mm以内的误差占93.75%。该方法对比传统方法精度有所提升,对比神经网络方法,所用训练样本少,训练过程更加简单。

机器人焊接技术在管道焊接中的应用现状

为介绍管道焊接中机器人焊接技术的应用情况,按照技术出现的顺序对油气管道焊接技术进行概括,分析了不同自动化焊接设备的工作原理,同时对工程中常用的设备进行了举例。文中介绍了管道铺设工程中相关的管道焊接检测技术以及管道接头腐蚀防护技术,最后介绍了焊接机器人的发展趋势,包括基于运动仿真的越障焊接、扩大应用场景与功能、引入新的传感器与原理。

基于机器视觉的智能建造H型钢焊接焊缝轨迹跟踪方法

为了有效提升智能建造H型钢焊接焊缝轨迹跟踪结果的准确性,提出一种基于机器视觉的智能建造H型钢焊接焊缝轨迹跟踪方法,通过机器视觉系统采集智能建造H型钢表面图像,采用非局部均值滤波和双边滤波相结合的方法进行降噪处理;对智能建造H型钢表面图像进行纹理分析,提取图像纹理特征,同时利用图像分割法确定焊缝区域;引入Shi-Tomasi算法进行角点检测,经过拟合直线求交点和逐行扫描获取焊缝坡口信息对应的拐点信息,确定焊缝中心,实现智能建造H型钢焊接焊缝轨迹跟踪。通过实际测试分析证明,所提方法可以准确跟踪焊缝轨迹,且具有较强的实时性,可以有效推进智能建造的发展。

基于改进ViT的熔池识别与焊接偏差在线检测方法

焊接偏差的精确检测是实现焊接机器人焊缝轨迹自动跟踪及智能化焊接的前提。提出了一种基于改进视觉转换器(Vision Transformer,ViT)的熔池识别与焊接偏差在线检测方法。首先,采用轻量级ViT模型Segformer作为基线模型,在其掩码分割前嵌入置换注意力(Shuffle Attention,SA)机制,以更好地捕获特征信息在空间和通道这2个维度中的依赖关系,从而提高模型的分割精度;其次,在多层感知机(Multilayer Perceptron,MLP)中加入上下文广播(Context Broadcasting,CB)模块,在保证模型低参数量的前提下提高泛化能力;最后,基于模型分割结果,提出一种焊接偏差计算方法来定量描述偏差检测精度。实验结果表明,相较于基线模型,所提出模型的平均交并比和平均像素准确率分别提高了2.67%和2.12%,且对于不同预设焊枪偏移情况均具有良好的泛化性,焊接偏差精度控制在±0.021 mm之内,为实现精密焊接焊缝跟踪提供基础。

基于视觉特征的机车钢结构焊接机器人焊缝跟踪方法研究

机车钢结构在智能焊接过程中,受到热输入、材料变形等因素的影响,焊缝的实际位置和形状会发生变化,在焊缝跟踪时,当前视觉帧间差方法容易忽略位置信息和焊缝轨迹位置偏差,导致跟踪误差大。对此,提出基于视觉特征的机车钢结构焊接机器人焊缝跟踪方法。选取以“棋盘格&条纹”混合图像作为基础,标定机车钢结构焊接机器人立体视觉位置,对焊接机器人激光条纹的中心线特征进行捕获;通过Roberts算子提取焊缝边缘信息,引入Hough变换确定焊缝中心线,实现焊缝识别,并将焊缝坐标展开转换,将其映射至机车钢结构焊接机器人的基础坐标系,在三维空间中构建焊缝轨迹;基于焊枪当前的位置信息和焊缝轨迹计算位置偏差,通过三次分均匀有理B样条对焊缝轨迹数据进行实时修正,实现机车钢结构焊接机器人焊缝实时跟踪。试验结果表明,所提方法可以有效提升机车钢结构焊接机器人焊缝跟踪精度。

基于图像处理技术的机器人焊接焊缝跟踪研究

为解决焊接机器人面对不同焊接工况时,焊缝跟踪精度低的问题,文章提出了基于图像处理技术的机器人焊接焊缝跟踪方法。利用CCD传感器采集机器人焊接焊缝图像,基于点操作的图像增强方法,灰度变换机器人焊接焊缝图像。利用SegFormer图像语义分割模型,通过Transformer编码器与多层感知机解码器,采用递归空洞自注意力机制,从机器人焊接焊缝的灰度图像中提取焊缝跟踪路径。利用卡尔曼滤波算法消除焊缝跟踪路径中的非焊接点,通过机器人焊接焊缝图像的坐标系转换,更新焊接焊缝轨迹堆栈,实现机器人焊接焊缝跟踪。实验结果表明,该方法能够依据机器人焊接焊缝图像处理结果,精准跟踪焊接焊缝,圆弧型与S型焊缝的跟踪误差低于3mm。

不锈钢氩弧焊焊管焊缝缺陷在线探伤系统研究

针对小管径不锈钢焊管在氩弧焊焊接过程中易发生焊缝偏移、焊缝位置存在缺陷等问题,研究了一种不锈钢焊管焊缝涡流在线探伤系统,该系统利用超声波对焊缝实现实时追踪,并结合涡流对焊缝缺陷进行检测,可以有效提高自动化检测精度,避免焊管缺陷。

焊接机器人在钢结构制造中的研究与应用

传统人工焊接存在自动化程度低、效率低下、质量参差不一等问题,而焊接机器人能大幅提高焊接效率,节省人力及改善焊接质量,保障同类焊缝的一致性,可被大规模应用于钢结构生产制造,因此,制造行业正逐渐向无人化、智能化方向转型。以油类垃圾箱为研究对象,采用Solidworks构建三维模型并进行静力学校核计算,利用Robot Stutio建立焊接工作站模拟仿真,对焊接轨迹进行离线编程,采用龙门式焊接机器人进行现场示教,调试激光纠偏系统,完成对油类垃圾箱的单片焊接,为焊接机器人在钢结构行业提供了应用依据。最后探讨了先进焊缝追踪技术和智能化焊接机器人,明确了焊接机器人未来应用的方向和趋势。

基于改进T-S模糊模型的弧焊机器人轨迹实时跟踪

针对弧焊机器人在实际作业中受到诸多不确定因素的影响,导致弧焊机器人焊接轨迹偏离预期轨迹的问题,提出一种基于改进T-S(Takagi-Sugeno)模糊模型的弧焊机器人轨迹实时跟踪方法。采用图像识别技术识别并定位焊缝,通过坐标转换,确定焊缝中心线在机器人坐标系中的位置坐标。利用蚁群算法根据焊缝中心线在机器人坐标系中的坐标,搜索从起始位置到目标位置的最优焊接路径。设计了一种改进区间二型T-S模糊模型,通过视觉传感器实时获取新的焊缝信息,计算轨迹误差,以此为输入,利用改进T-S模糊模型对机器人的运动轨迹进行实时调整。实验结果表明,采用改进T-S模糊模型后,机器人轨迹跟踪离散度均降至1 mm以下,弧焊机器人轨迹与预期轨迹高度重合,比传统方法跟踪离散度大大降低。

高频列车荷载作用下板式无砟轨道垂向应力衰减特性

为了优化无砟轨道结构设计并为确定动力系数取值提供参考,以CRTSⅠ型板式无砟轨道为研究对象,采用LS-DYNA软件建立车辆-轨道耦合动力学模型,分析了不同速度级时车轮扁疤和焊缝激励下的轮轨垂向力时频特征,阐明了轮轨垂向力随车速的变化规律;进而分析了轨道结构垂向应力沿深度的变化规律,计算了轨道结构动力系数并揭示了动力系数沿轨道结构深度的变化规律。结果表明:车速在400 km/h内,车轮扁疤不平顺作用下轮轨垂向力主频大于500 Hz,而焊缝不平顺作用下轮轨垂向力存在两个主频,较高的主频大于500 Hz,较低的主频处于(49.9±10)Hz内;高频列车荷载下的轨道垂向应力沿着轨道结构深度衰减明显,轨道板底部的垂向应力约为轨道板顶部的15%,底座板底部的垂向应力约为轨道板顶部的4%;轨道结构动力系数沿轨道结构深度显著降低,在钢轨至轨道板之间衰减约40%,底座板底部动力系数比钢轨上减小约50%。建议适当降低高频列车荷载作用下无砟轨道设计和检算动力系数。

船舱格子型角焊缝移动焊接机器人系统设计

针对大型船舱格子型角焊缝空间狭小,大型自动化设备与工人难以高质量完成焊接作业的现状,设计了一种可进行连贯直角角焊缝焊接的移动焊接机器人系统。该机器人由机械结构与电路控制系统组成,其中机械结构包括双轴后驱车体、可伸缩接触式传感器、机器人升降旋转复合机构、焊枪位姿调整机构;电控系统硬件包括PLC控制器、行程限位开关、步进电机与配套驱动器、直流减速电机和配套的运动状态调控板。另外,还建立了机器人在直角段角焊缝焊接过程的运动学模型,获取了该段焊缝在均分为10段后每段焊枪的伸缩速度与机器人旋转角速度的数值,为直角型角焊缝焊接的跟踪控制过程提供了PLC编程的数据基础;并基于可伸缩接触式传感器,实现了直线型角焊缝的路径跟踪。实验证明,该机器人能实现小空间直角型角焊缝与直线型角焊缝的连贯焊接,焊缝成型质量好,无焊穿、气孔、焊瘤、咬边、裂纹等不良状况。

Welding Deviation Detection Algorithm Based on Extremum of Molten Pool Image Contour

The welding deviation detection is the basis of robotic tracking welding, but the on-line real-time measurement of welding deviation is still not well solved by the existing methods. There is plenty of information in the gas metal arc welding（GMAW） molten pool images that is very important for the control of welding seam tracking. The physical meaning for the curvature extremum of molten pool contour is revealed by researching the molten pool images, that is, the deviation information points of welding wire center and the molten tip center are the maxima and the local maxima of the contour curvature, and the horizontal welding deviation is the position difference of these two extremum points. A new method of weld deviation detection is presented, including the process of preprocessing molten pool images, extracting and segmenting the contours, obtaining the contour extremum points, and calculating the welding deviation, etc. Extracting the contours is the premise, segmenting the contour lines is the foundation, and obtaining the contour extremum points is the key. The contour images can be extracted with the method of discrete dyadic wavelet transform, which is divided into two sub contours including welding wire and molten tip separately. The curvature value of each point of the two sub contour lines is calculated based on the approximate curvature formula of multi-points for plane curve, and the two points of the curvature extremum are the characteristics needed for the welding deviation calculation. The results of the tests and analyses show that the maximum error of the obtained on-line welding deviation is 2 pixels（0.16 ram）, and the algorithm is stable enough to meet the requirements of the pipeline in real-time control at a speed of less than 500 mm/min. The method can be applied to the on-line automatic welding deviation detection.