公开(公告)号：KR1020120045821A

公开(公告)日：2012-05-09

申请号：KR1020100107639

申请日：2010-11-01

Applicant: 대우조선해양 주식회사

Inventor： 김재형 ,  김병만 ,  민지호

IPC: B23K37/00 ,  B23K9/127 ,  B25J9/16 ,  B25J9/22

Abstract: PURPOSE: A calibration method between a robot and an LVS(Laser Vision System) using a lug is provided to improve the speed and accuracy of work by locating a robot in a set initial position by a worker so that the robot automatically implement successive processes. CONSTITUTION: A calibration method between a robot and an LVS(102) using a lug(1) comprises the steps of: locating a robot in the initial position of the lug, moving the robot to a measurement position of a touch sensor(101), implementing three-point measurement by the touch sensor in order to measure boundary feature points of the lug, moving the robot to a measurement position of the LVS, setting the joint type of the LVS into a simple fellet shape, and continuously measuring the feature points using the LVS by horizontally moving the robot along the boundary of the lug.

Abstract translation: 目的：提供使用凸耳的机器人和LVS（激光视觉系统）之间的校准方法，以通过将机器人定位在工作人员设定的初始位置来提高工作的速度和精度，从而使机器人自动执行连续的过程。 构成：使用凸耳（1）的机器人和LVS（102）之间的校准方法包括以下步骤：将机器人定位在凸耳的初始位置，将机器人移动到触摸传感器（101）的测量位置， 通过触摸传感器实现三点测量，以测量凸耳的边界特征点，将机器人移动到LVS的测量位置，将LVS的关节类型设置为简单的绒毛形状，并连续测量特征 通过沿着凸缘的边界水平移动机器人，使用LVS的点。

公开(公告)号：KR1020120036145A

公开(公告)日：2012-04-17

申请号：KR1020100097860

申请日：2010-10-07

Applicant: 대우조선해양 주식회사

Inventor： 김병만 ,  민지호 ,  김재형

IPC: G01B11/24 ,  G01B11/25 ,  G06T7/60 ,  G06T7/00 ,  B23K31/12

CPC classification number: G01B11/24 ,  B23K31/12 ,  G01B11/2518 ,  G06T7/521 ,  G06T7/60

Abstract: PURPOSE: A detecting method for a characteristic point of a welding joint using a laser vision system is provided to automatically detect a characteristic point of a welding joint without an additional setting by a user when detecting the characteristic point of the welding joint from measured welding joint image. CONSTITUTION: A detecting method for a characteristic point of a welding joint using a laser vision system is as follows. An image of a welding object obtained through a laser vision system is binarized(S120). An edge detecting algorithm is applied to the image binarized(S130). Candidate points where the characteristic point can generate are detected(S130). The center parts of the candidate points detected are detected(S140). The binarized images of surroundings around the detected candidate points are compared(S150). Whether or not the detected candidate points are the characteristic point is decided(S160).

Abstract translation: 目的：提供一种使用激光视觉系统的焊接接头的特征点的检测方法，用于当从测量的焊接接头检测到焊接接头的特征点时，自动检测焊接接头的特征点，而无需用户的额外设置 图片。 构成：使用激光视觉系统的焊接接头的特征点的检测方法如下。 通过激光视觉系统获得的焊接对象的图像被二值化（S120）。 边缘检测算法应用于二值化图像（S130）。 检测特征点可以产生的候选点（S130）。 检测出候选点的中心部分（S140）。 对检测到的候选点周围环境的二值化图像进行比较（S150）。 确定检测到的候选点是否是特征点（S160）。