Process planning for additive/subtractive solid freeform fabrication using medial axis transform

Additive/Subtractive Solid Freeform, Fabrication (SFF) integrates material addition (deposition) and removal (machining) to build up three-dimensional objects incrementally. This class of processes offers sophisticated design flexibility with engineering materials, three-dimensional layer building, and the ability to fabricate complex engineering devices and multi-material objects. However, planning for such processes exhibits rigorous challenges due to process flexibility and highly demanding planning automation. These challenges, moreover, can not be sufficiently tackled via common boundary representation of geometric models. In this thesis, various techniques based on the Medial Axis Transform (MAT) are presented. The medial axis transform encodes intrinsic shape characteristics into a lower dimensional metric. The MAT together with the boundary representation empowers shape manipulation and geometric reasoning. Although numerous algorithms have been proposed to recognize the MAT of polygonal objects, a robust model for arbitrarily shaped regions, especially suitable for engineering designs, is still an art of research. The approaches presented in this thesis utilize representation of the MAT in terms of clearance functions on the object boundary. The clearance functions are computed via a divide-and-conquer methodology. Various planning techniques based on the proposed MAT representation are developed to tackle three process planning problems in additive/subtractive SFF. First, an automated manufacturability analysis approach is presented to assist in evaluation of part build sequences. Such a method allows fast identification of feasible build sequences that permit cutting tool access at all build stages. Second, a shape optimization scheme is developed to compute optimal layer geometry for material deposition. The shape is so optimized that the connected and smooth deposition paths can be generated. Third, an efficient cutter selection strategy is proposed for shaping near-net deposition as well as for bulk material removal. A procedure based on the histograms of shape thickness is presented to efficiently compute an optimal set of cutters for achieving the minimal machining time. The proposed MAT representation and planning approaches apply not only to additive/subtractive solid freeform fabrication but also to various conventional manufacturing processes. The potential of utilizing such techniques for geometric reasoning and process planning is yet to be explored.