研究目的
To improve the efficiency of welding robots by proposing a novel three-dimensional(3D) path extraction method of weld seams based on stereo structured light sensor.
研究成果
The proposed 3D path teaching method based on stereo structured light vision system could improve the efficiency of the teaching-playback mode and the off-line programming mode of welding robot. The measurement resolution is less than 0.7 mm and the proposed method is suitable for V-type butt joint before welding.
研究不足
The proposed method is only suitable for the V-type butt joint and the measurement precision of the proposed method needs to be improved.
1：Experimental Design and Method Selection:
The stereo structured light vision is used to accomplish the task of 3D reconstruction in the industrial environment. It could realize fast and high-precision 3D reconstruction of welding work piece.
2：Sample Selection and Data Sources:
The straight and curve V-type butt joints are set as experimental objects.
3：List of Experimental Equipment and Materials:
The Motoman UP6 industrial manipulator is used to do the 3D seam path extraction. The high-performance CMOS industrial camera with a total pixel number of 2592*1944 MER-500-7UC is adopted.
4：Experimental Procedures and Operational Workflow:
The optical projector projects the fringe patterns to the surface of the welding work piece and the industrial cameras capture the images with distorted fringe patterns modulated by the surface height of the welding work piece. Then these images are transferred into the vision computer through the USB bus. Finally, the 3D path extraction of weld seam could be realized by a series of vision algorithms and calibration algorithms.
5：Data Analysis Methods:
The root mean square error(RMSE) of the fitting results is computed to verify the proposed method.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
