A novel path planning method of robotic grinding for free-form weld seam based on 3D point cloud

In robot grinding, the path planning has always been the main factor affecting the grinding efficiency. To improve the accuracy and automation level of robot grinding, a novel path planning method based on workpiece point cloud is proposed for the grinding of weld seam on curved surface. First, the point cloud of workpiece is obtained by the binocular structured light camera. After data preprocessing, this paper presents a novel approach of tangent planes determination based on the idea of point cloud slicing, which involves 3D projection and image binary extraction. Second, an approach for extracting the weld feature points is proposed based on the deviation term. By calculating the intersecting points between tangent planes and workpiece point cloud, this approach identifies the weld profile and extracts the feature points located in weld center. Then, to reduce the vibration in grinding, the feature points are polynomial fitted to generate a continuous weld grinding path, and a method of posture planning for grinding tool is presented based on grinding tool model and grinding process. Finally, this paper builds the “robot + 3D vision” platform and designs the grinding actuator, the effectiveness of this proposed method is verified by experiments.