提出了基于条纹式激光传感器的机器人焊缝跟踪系统，采用机器人末端安装条纹激光传感器，在焊接过程中实时获取焊缝轮廓数据；利用小波变换模极大值理论对焊缝轮廓数据进行分析，确定焊缝特征点；以特征点为分界点，采用最小二乘法拟合焊缝直线并求取交点，精确确定焊缝轮廓特征点坐标；通过机器人坐标变换实时获得焊缝运动轨迹，从而实现焊缝跟踪。对不同的V 型焊缝进行实验验证，并对有无采用最小二乘法精确定位进行对比实验。实验结果表明：只采用小波模极大值理论获得的跟踪误差约为0.514 mm,但轨迹存在较严重的抖动现象；而采用了最小二乘法精确定位特征点后，跟踪误差约为0.304 mm，抖动现象明显减弱,满足自动焊接误差要求在0.5 mm之内的要求。

A new seam-tracking system based on stripe type laser sensor and welding robot is proposed. The laser sensor is fixed to the end of the robot. The weld propfile data is abstracted in real- time during the welding process. And the wavelet transform modulus maxima theory is used to extract the welding characteristic- points. The contour data is grouped based on the characteristic- points. And then, the least square method is used to fit the weld lines and calculate the intersection point of straight line, which is aimed to determine the coordinates of the characteristic- points accurately. At last, the coordinate transformation calculation is used to get the trajectory of the welding torch in real-time. Then, two experimental verifications for different V- shape welds are carried on. Comparing experiments between only using wavelet transform modulus maxima theory and using wavelet transform modulus maxima and the least square method are performed. The experimental results show that the error of the former is about 0.514 mm, but the trajectory exists serious jitter phenomenon. The error of the other is about 0.304 mm and the jitter phenomenon significantly diminishes, which satisfies the requirements of automatic welding of requirements within 0.5 mm.