Prismatic CAD Model Generation via ML Autoencoder
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer-aided design (CAD) software often generates 3D models from generative design processes that are not directly suitable for manufacturing, requiring manual adjustments to address issues like rounded edges and unconventional shapes.
Innovation Solution
A method using machine learning, specifically 2D and 3D autoencoders, to process input embeddings and generate fitted parametric sketch models, which can be extruded into boundary representation (B-Rep) models suitable for manufacturing, by training on voxel models and signed distance field images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If generative design process is used to automatically generate 3D geometry, then productivity is improved, but manufacturing suitability deteriorates
Solution Approach 1:
A machine learning model acts as an intermediary between the generative design process and manufacturing. The model takes voxel models from generative design as input and outputs prismatic B-Rep models suitable for manufacturing, automatically translating between the two representations without manual intervention.
Solution Approach 2:
The patent replaces the manual mechanical process of adjusting 3D models with an automated machine learning system. The ML model automatically converts voxel models to prismatic B-Rep models, substituting human operators with an intelligent algorithm that performs the transformation.
2Ease of manufacture
If manual adjustments are made to generate modified 3D models, then manufacturing suitability is improved, but loss of time increases
Solution Approach 1:
The system performs self-service by automatically converting voxel models to manufacturable B-Rep models without requiring human operators. The machine learning model independently completes the transformation task that previously required manual intervention, eliminating the time loss associated with human adjustments.
Solution Approach 2:
Manual adjustment operations are replaced with an automated machine learning system that performs the conversion from voxel to B-Rep format. This substitution eliminates the time-consuming manual process while maintaining manufacturing suitability.
3Productivity
If voxel models are used directly from generative design, then productivity is improved, but manufacturing precision deteriorates
Solution Approach 1:
The machine learning model serves as an intermediary transformation layer that converts voxel models to prismatic B-Rep models. This intermediary process maintains the productivity benefit of automated generation while improving manufacturing precision by outputting models in the appropriate format with correct geometric properties.
Solution Approach 2:
The patent changes the geometric parameters and representation format of the model during conversion. Voxel models with their discrete grid structure are transformed into B-Rep models with continuous surfaces and precise boundaries, changing the fundamental parameters of the geometric representation to achieve manufacturing precision.
Data Source
AI summary
Methods, systems, and apparatus, including medium-encoded computer program products, for computer aided design and manufacture of physical structures by generating prismatic CAD models using machine learning, include: obtaining an input embedding that encodes a representation of a target two-dimensional (2D) shape; processing the input embedding using a 2D decoder of a 2D autoencoder to obtain a decoded representation of the target 2D shape; determining a fitted 2D parametric sketch model for the input embedding, including: finding a 2D parametric sketch model for the input embedding using a search in an embedding space of the 2D autoencoder and a database of sketch models associated with the 2D autoencoder, and fitting the 2D parametric sketch model to the decoded representation of the target 2D shape; and using the fitted 2D parametric sketch model in a computer modeling program.