卡尔曼滤波焊缝跟踪控制

Seam tracking control using a Kalman filter

中文关键词  焊缝跟踪  卡尔曼滤波  熔池图像质心

英文关键词  seam tracking  Kalman filtering  weld pool image centroid

中文摘要

      在视觉传感的电弧自动焊接过程中, 需要根据视觉信息来控制电弧准确地跟踪焊缝. 由于强烈的弧光干扰, 使得从焊接区图像中直接提取电弧与焊缝的偏差信息十分困难. 为此提出一种利用熔池图像质心和卡尔曼滤波来间接获取电弧与焊缝偏差的方法. 选择熔池图像质心作为状态向量, 建立基于图像质心的状态方程和焊缝位置测量方程. 利用卡尔曼滤波消除过程噪声和测量噪声的影响, 通过对熔池图像质心的状态估计, 准确获取焊缝位置以及电弧与焊缝之间的偏差量, 为自动焊接过程的焊缝跟踪控制提供准确信息. 焊接试验结果表明, 利用卡尔曼滤波方法可有效降低过程噪声和测量噪声的影响, 从而提高焊缝跟踪控制精度.

英文摘要

      During an automatic arc welding process which visual sensors are used, the electric arc should be controlled to track the weld seam accurately based on the visual sensing information. However, it is very difficult to acquire the deviation between the electric arc and the weld seam from the welding images because of strong arc disturbance. An approach is proposed that this deviation can be obtained by using the weld pool image centroid and a Kalman filter. The centroid of the weld pool image from a visual sensor is extracted as the measurement eigenvector. Based on the weld pool image centroid, the state equation and the weld position measurement equation are established. The advantages of a Kalman filter is taken to reduce the error of weld detection caused by the process and the measurement noises. Also, the weld position and the deviation between the electric arc and the weld seam can be estimated by using the estimation of the weld pool image centroid state. It can provide the accurate information for seam tracking control during the automatic welding process. Actual welding experimental results have demonstrated the effectiveness that the influence of the process and measurement noises can be reduced and the seam tracking accuracy can be improved by using the Kalman filtering technique.