Session: CIE-24-01 - AMS-CAPPD-SEIKEM: Artificial Intelligence and Machine Learning in Design and Manufacturing

Paper Number: 89307

89307 - Manufacturing Process Classification Based on Distance Rotationally Invariant Convolutions 

  Given a design part, the task of manufacturing process classification identifies an appropriate manufacturing process to fabricate it. Our previous research proposed a large dataset for manufacturing process classification and achieved accurate results based on a combination of a convolutional neural network (CNN) and the heat kernel signature (HKS) for triangle mesh. In this paper, we constructed a classification method based on rotation-invariant shape descriptors and a neural network for point clouds, and it achieved better accuracy than all previous methods. This method uses a point cloud part representation, in contrast to the triangle mesh representation used in our previous work. The first step extracted rotation-invariant features consisting of a set of distances between points in the point cloud. Then, the extracted shape descriptors were fed into a CNN for the classification of manufacturing processes. In the end, the trained neural network achieves the properties of translation invariance, rotation invariance, and scale invariance. In addition, we provided two visualization methods for interpreting the intermediate layers of the neural network. Though manufacturing processes selection depends on other parameters like material and tolerance, the focus of this paper lies on part shape. Our work demonstrates that part shape attributes alone are adequate for discriminating between different manufacturing processes considered.

Presenting Author: David Rosen Georgia Institute of Technology

Presenting Author Biography: David Rosen is a Professor in the School of Mechanical Engineering at the Georgia Institute of Technology, where he is Director of the Rapid Prototyping & Manufacturing Institute. Additionally, he is the Research Director of the Digital Manufacturing & Design Centre at the Singapore University of Technology & Design. His research interests include computer-aided design, additive manufacturing (AM), and design methodology. He is a Fellow of ASME and chairs the ASTM F42 subcommittee on design for additive manufacturing standards. Additionally, he is a co-author of a leading textbook on AM.

Authors:

Zhichao Wang Georgia Institute of Technology
David Rosen Georgia Institute of Technology