Adaptive Toolpath Planning for Hybrid Manufacturing Based on Raw 3D Scanning Data

The wide industrial adoption of metal Additive Manufacturing (AM) technologies has shown that AM can rarely thrive on its own but should be integrated in a wider manufacturing chain. The integration of AM with milling, to form the so-called hybrid manufacturing process is one of the most popular approaches. A key bottleneck of hybrid manufacturing is toolpath planning. In a hybrid manufacturing workflow, 3D scanning is often integrated between milling and AM, in an inspection step, which is highly important for the correct planning of the next process, since AM parts can have a significant divergence from their original CAD. In order to plan the toolpath of the successive AM or milling process, one must consider the 3D scan as an input, posing a new challenge compared to traditional toolpath planning technologies. The surface reconstruction of the part, based on the point cloud can be very computationally heavy. To this end, new techniques for toolpath planning directly from the raw 3D scanning data are necessary. This study presents such a technique to provide an automated tool for toolpath planning of both AM and machining processes, based on point cloud data. Edge extraction is performed through eigenvalue analysis, followed by segmentation of the part based on popular clustering algorithms. Next, curve fitting on the external and internal parts of the segmented point cloud is performed using alpha shapes, followed by polyline offsetting for the toolpath generation. The methodology is validated through a case study on a real component.