Study on Intelligence of Welding Robot with Visual Sensing System

视觉传感系统焊接机器人智能化研究

基本信息

批准号：
07650852
负责人：
SUGA Yasuo
金额：
$ 1.6万
依托单位：
Keio University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
1995
资助国家：
日本
起止时间：
1995 至 1996
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-07650852/
关键词：
Welding robot Visual sensing Welding line Automatic seam tracking Molten pool Control of penetration Image processing 溶融池視覚センサ知能化溶接制御シームトラッキングニューラルネットワーク

项目摘要

In last year, a basic system for recognizing welding line by visual sensing was constructed. In this year, a robust control system for recognizing and tracking the welding line in butt welding of thin steel plates was constructed. In this system, the recognition of the welding line was performed by applying the image data acquired by CCD camera to artificial neural network (N.N.) and position and shape of the welding line was detected. Furthermore, in order to obtain sound weld regardless change of welding conditions, such as groove shape, gap of groove and so on, system to diagnose the welding phenomena during welding and to control welding conditions were constructed.Main results obtained are summarized as follows.(1) Preprocessing of image : Before inputting image data into N.N., smoothing was applied to the image using 2-D space filter to eliminate noise. Then the LOG filter was applied to it in order to enhance the edge of the welding line image and the binarisation was applied. T … More he image data obtained were used as the input data to N.N..(2) Tracking experiments by the developed system : Experiments for recognizing welding lines of base metals which have several surface conditions, type of material, groove shape, type of groove, lighting conditions and arc conditions were performed, and the effectiveness of the recognizing system constructed was confirmed. Furthermore, butt welding experiments of mild steel plates were performed and effectiveness of the system to automatic welding was confirmed.(3) Monitoring of mlten pool : In order to realize real-time control of welding condition and obtain a sound weld, a monitoring system of molten pool in arc welding was constructed. As the relationship between the molten pool shape and penetration of weld has intimate relationship, if the shape of the molten pool can control, the sound weld with optimum penetration may be obtained. In this study a monitoring system and arc current control system were combined and a basic welding control system was constructed. Less

去年，构建了一种通过视觉传感识别焊接线的基本系统。在今年，构建了一个可靠的控制系统，用于识别和跟踪细钢板的对接焊接中的焊接线。在此系统中，通过将CCD摄像机获取的图像数据应用于人工神经元网络（N.N.）以及检测到焊接线的位置和形状来执行焊接线的识别。此外，为了获得声音焊接，不管焊接条件的变化，例如凹槽形状，凹槽间隙等，构建焊接过程和控制焊接条件的系统以诊断焊接现象的构建。所获得的结果。（按以下方式汇总了图像的预先启动图像。噪音。然后将对数滤波器应用于其，以增强焊接线图像的边缘并应用二线化。 t…获得的更多HE图像数据用作N.N.的输入数据。（2）通过已开发的系统跟踪实验：用于识别具有多种表面条件，材料类型，凹槽形状的类型，凹槽，照明条件和ARC条件的焊接线的实验，并确认了构建构建系统的有效性。此外，进行了低碳钢板的对接焊接实验，并确认系统对自动焊接的有效性。（3）监测MLTEN池：为了实现对焊接条件的实时控制，并获得了声音焊接的实时控制，并构建了ARC焊接中熔融池的监测系统。由于熔融池形状和焊缝穿透之间的关系具有亲密关系，如果熔融池的形状可以控制，则可以获得具有最佳穿透性的声焊接。在这项研究中，组合了监测系统和电弧电流控制系统，并构建了基本的焊接控制系统。较少的