An official website of the United States government
Abstract This paper presents NURBS-OT (non-uniform rational B-splines—optimal transport), a new approach in the field of computer graphics and computer-aided design (CAD)/computer-aided manufacturing (CAM) for modeling complex free-form designs like aerodynamic and hydrodynamic structures, traditionally shaped by parametric curves such as Bézier, B-spline, and NURBS. Unlike prior models that used generative adversarial networks (GANs) involving large and complex parameter sets, our approach leverages a much lighter (0.37M versus 5.05M of BézierGAN), theoretically robust method by blending optimal transport with NURBS. This integration facilitates a more efficient generation of curvilinear designs. The efficacy of NURBS-OT has been validated through extensive testing on the University of Illinois Urbana-Champaign (UIUC) airfoil and superformula datasets, where it showed enhanced performance on various metrics. This demonstrates its ability to produce precise, realistic, and esthetically coherent designs, marking a significant advancement by merging classical geometrical techniques with modern deep learning.
more »
« less
VITRUVION: A GENERATIVE MODEL OF PARAMETRIC CAD SKETCHES
Parametric computer-aided design (CAD) tools are the predominant way that engineers specify physical structures, from bicycle pedals to airplanes to printed circuit boards. The key characteristic of parametric CAD is that design intent is encoded not only via geometric primitives, but also by parameterized constraints between the elements. This relational specification can be viewed as the construction of a constraint program, allowing edits to coherently propagate to other parts of the design. Machine learning offers the intriguing possibility of accelerating the de- sign process via generative modeling of these structures, enabling new tools such as autocompletion, constraint inference, and conditional synthesis. In this work, we present such an approach to generative modeling of parametric CAD sketches, which constitute the basic computational building blocks of modern mechanical design. Our model, trained on real-world designs from the SketchGraphs dataset, autoregressively synthesizes sketches as sequences of primitives, with initial coordinates, and constraints that reference back to the sampled primitives. As samples from the model match the constraint graph representation used in standard CAD software, they may be directly imported, solved, and edited according to down- stream design tasks. In addition, we condition the model on various contexts, including partial sketches (primers) and images of hand-drawn sketches. Evaluation of the proposed approach demonstrates its ability to synthesize realistic CAD sketches and its potential to aid the mechanical design workflow.
more »
« less
- Award ID(s):
- 2118201
- PAR ID:
- 10339655
- Date Published:
- Journal Name:
- The International Conference on Learning Representations (ICLR)
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Computer-aided design (CAD) is a standard design tool used in engineering practice and by students. CAD has become increasingly analytic and inventive in incorporating artificial intelligence (AI) approaches to design, e.g., generative design (GD), to help expand designers' divergent thinking. However, generative design technologies are relatively new, we know little about generative design thinking in students. This research aims to advance our understanding of the relationship between aspects of generative design thinking and traditional design thinking. This study was set in an introductory graphics and design course where student designers used Fusion 360 to optimize a bicycle wheel frame. We collected the following data from the sample: divergent and convergent psychological tests and an open-ended response to a generative design prompt (called the generative design reasoning elicitation problem). A Spearman's rank correlation showed no statistically significant relationship between generative design reasoning and divergent thinking. However, an analysis of variance found a significant difference in generative design reasoning and convergent thinking between groups with moderate GD reasoning and low GD reasoning. This study shows that new computational tools might present the same challenges to beginning designers as conventional tools. Instructors should be aware of informed design practices and encourage students to grow into informed designers by introducing them to new technology, such as generative design.more » « less
-
Generative models such as diffusion models, excel at capturing high-dimensional distributions with diverse input modalities, e.g. robot trajectories, but are less effective at multistep constraint reasoning. Task and Motion Planning (TAMP) approaches are suited for planning multi-step autonomous robot manipulation. However, it can be difficult to apply them to domains where the environment and its dynamics are not fully known. We propose to overcome these limitations by composing diffusion models using a TAMP system. We use the learned components for constraints and samplers that are difficult to engineer in the planning model, and use a TAMP solver to search for the task plan with constraint-satisfying action parameter values. To tractably make predictions for unseen objects in the environment, we define the learned samplers and TAMP operators on learned latent embedding of changing object states. We evaluate our approach in a simulated articulated object manipulation domain and show how the combination of classical TAMP, generative modeling, and latent embedding enables multi-step constraint-based reasoning. We also apply the learned sampler in the real world.more » « less
-
Deep generative models of 3D shapes have received a great deal of research interest. Yet, almost all of them generate discrete shape representations, such as voxels, point clouds, and polygon meshes. We present the first 3D generative model for a drastically different shape representation—describing a shape as a sequence of computer-aided design (CAD) operations. Unlike meshes and point clouds, CAD models encode the user creation process of 3D shapes, widely used in numerous industrial and engineering design tasks. However, the sequential and irregular structure of CAD operations poses significant challenges for existing 3D generative models. Drawing an analogy between CAD operations and natural language, we propose a CAD generative network based on the Transformer. We demonstrate the performance of our model for both shape autoencoding and random shape generation. To train our network, we create a new CAD dataset consisting of 178,238 models and their CAD construction sequences. We have made this dataset publicly available to promote future research on this topic.more » « less
-
Abstract This work presents a method for the topology optimization of welded frame structures to minimize the manufacturing cost. The structures considered here consist of assemblies of geometric primitives such as bars and plates that are common in welded frame construction. A geometry projection technique is used to map the primitives onto a continuous density field that is subsequently used to interpolate material properties. As in density-based topology optimization techniques, the ensuing ersatz material is used to perform the structural analysis on a fixed mesh, thereby circumventing the need for re-meshing upon design changes. The distinct advantage of the representation by geometric primitives is the ease of computation of the manufacturing cost in terms of the design parameters, while the geometry projection facilitates the analysis within a continuous design region. The proposed method is demonstrated via the manufacturing-cost-minimization subject to a displacement constraint of 2D bar, 3D bar, and plate structures.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
