X. Zhen, J. C. Y. Seng and N. Somani, "Adaptive Automatic Robot Tool Path Generation Based on Point Cloud Projection Algorithm," 2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA), Zaragoza, Spain, 2019, pp. 341-347. doi: 10.1109/ETFA.2019.8869301
Abstract:
Many industry manufacturing processes require a lot of manpower to accomplish tasks manually, for example, manual polishing and masking. Industrial robot can be used to replace most of the tedious and repeated tasks. However, using robot program to generate the tool path for the manufacturing process might need programming skills and expertise. Besides, Computer Aided Design (CAD) files might not be available or accurate for the engineer to design the robot tool path. Hence, we propose an automatic way to generate the adaptive robot tool path for manufacturing process by using scan point cloud data of the target coupon. The core algorithm is based on point cloud projection on plane, tool path pattern design and reverse transform matrix to project the 2d tool path back to 3d point cloud. The algorithm is based on Point Cloud Library (PCL) and OpenCV libraries. After the toolpath is generated in the point cloud, Robot Operating System (ROS) is used to plan trajectory and check for collision. The automated tool path generation algorithm can be applied to multiple manufacturing process, such as masking, polishing and painting.