Abstract
One of the advantages additive manufacturing (AM) has over traditional manufacturing techniques is the complexity of the parts it can produce. However, this complexity creates a challenge in quantifying the geometric deviation between the desired and printed shapes due to the lack of clear reference features on the printed part. Existing methods for quantifying shape deviation in parts made via AM make simplifying assumptions about reference features, thus limiting their versatility. To address this shortcoming, in this work, the iterative closest point (ICP) algorithm is adopted to achieve accurate registration between desired and printed shapes. To enhance its versatility, the proposed method makes no assumptions about the reference features. Instead, it determines the best reference features automatically by performing ICP using a weighted sliding window. To validate the proposed method, two simulated case studies are used. In both cases, the proposed method is able to automatically determine reference features, hence accurately represent geometric deviation, without need for simplifying assumptions.
