Machine-assisted part design with automated design request interpretation

Patent metadata

Field	Value
Publication number	US-10908587-B2
Application number	US-201715796074-A
Country	US
Kind code	B2
Filing date	Oct 27, 2017
Priority date	Oct 27, 2016
Publication date	Feb 2, 2021
Grant date	Feb 2, 2021

Abstract

A factory server receives part requests from customer devices and controls one or more manufacturing tools, such as 3D printers, to fabricate the requested parts. The factory server implements several features to streamline the process of fabricating parts using the manufacturing tools. For instance, the factory server can facilitate the design of a part by extracting features from the part request and identifying model files having those features. The factory server can also select an orientation in which to fabricate the part and determine print settings to use when fabricating the part. In addition, the factory server can implement a process to fabricate a three-dimensional part with a two-dimensional image applied to one or more of its external surfaces. Furthermore, the factory server can also generate a layout of multiple part instances on a build plate of a 3D printer so that multiple part instances can be fabricated at once.

First claim

What is claimed is:

1. A method comprising:
receiving, from a customer device associated with a customer, a design request for a part to be fabricated by an additive manufacturing process, the design request including descriptive information describing preferred properties of the part and an environment and use of the part, the descriptive information comprising a written description of the part and one or more images of the part, wherein the descriptive information describing the environment and use of the part includes mechanical loads and temperatures ranges the part will be exposed to and desired material properties of the part;
extracting one or more features from the descriptive information, each of the extracted features representing a desired property of the part, wherein the extracting comprises:
identifying, using image processing, one or more structures within each of the one or more images of the part; and
mapping, using a machine learning model, the identified structures to one or more keywords;

wherein each of the extracted features is a scalar value representing a probability that the descriptive information contains the property represented by the feature;
selecting one or more three-dimensional (3D) model files for display to a designer, each of the 3D model files stored in a model file library and tagged with one or more features, the selection of the 3D model files performed based on a measure of similarity between the extracted features and the tagged features for each of the 3D model files;
wherein the measure of similarity between the extracted features and the tagged features for a 3D model file is a cosine similarity between a feature vector comprising the extracted features and a feature vector comprising the tagged features for the 3D model file;
sending the selected model files to a factory client device to be displayed to the designer;
receiving a new 3D model file for the part from the designer, the new 3D model file including a portion of one of the one or more selected 3D model files and new portions created by the designer;
generating manufacturing instructions for the part with the new model file, the manufacturing instructions including a layout of one or more instances of the part on a build plate;
determining print settings based on the part's environment and use, the print settings including a material type for the part;
sending the new model file, print settings, and manufacturing instructions to a printer.

2. The method of claim 1 , further comprising:
before selecting the one or more 3D model files for sending to the factory client device, identifying a plurality of candidate 3D model files stored in the model file library based on a measure of similarity between the extracted features and the tagged features for each of the candidate 3D model files;
sending the plurality of candidate 3D model files to the customer device to be displayed to the customer; and
receiving, from the customer device, a selection by the customer of one or more of the candidate 3D model files, wherein the selection of the one or more 3D model files for sending to the factory client is performed based on a measure of similarity between the modified extracted features and the tagged features for each of the 3D model files.

3. The method of claim 1 , wherein the descriptive information does not comprise a 3D model file of the part.

4. The method of claim 1 , wherein extracting one or more features from the descriptive information further comprises:
identifying, using natural language processing, one or more keywords from the written description of the part.

5. The method of claim 1 , wherein at least one of the extracted features is a binary value having a value selected from the group consisting of: a first value representing presence of the property represented by the feature and a second value representing absence of the property represented by the feature.

6. A non-transitory computer-readable storage medium having computer-executable instructions stored thereon, the instructions when executed by a processor causing the processor to perform operations comprising:
receiving, from a customer device associated with a customer, a design request for a part to be fabricated by an additive manufacturing process, the design request including descriptive information describing preferred properties of part and an environment and use of the part, the descriptive information comprising a written description of the part and one or more images of the part, wherein the descriptive information describing the environment and use of the part includes mechanical loads and temperatures ranges the part will be exposed to and desired material properties of the part;
extracting one or more features from the descriptive information, each of the extracted features representing a desired property of the part, wherein the extracting comprises:
identifying, using image processing, one or more structures within each of the one or more images of the part; and
mapping, using a machine learning model, the identified structures to one or more keywords;

wherein each of the extracted features is a scalar value representing a probability that the descriptive information contains the property represented by the feature;
selecting one or more three-dimensional (3D) model files for display to a designer, each of the 3D model files stored in a model file library and tagged with one or more features, the selection of the 3D model files performed based on a measure of similarity between the extracted features and the tagged features for each of the 3D model files;
wherein the measure of similarity between the extracted features and the tagged features for a 3D model file is a cosine similarity between a feature vector comprising the extracted features and a feature vector comprising the tagged features for the 3D model file;
sending the selected model files to a factory client device to be displayed to the designer;
receiving a new 3D model file for the part from the designer, the new 3D model file including a portion of one of the one or more selected 3D model files and new portions created by the designer;
generating manufacturing instructions for the part with the new model file, the manufacturing instructions including a layout of one or more instances of the part on a build plate;
determining print settings based on the part's environment and use, the print settings including a material type for the part;
sending the new model file, print settings, and manufacturing instructions to a printer.

7. The non-transitory computer-readable storage medium of claim 6 , the operations further comprising:
before selecting the one or more 3D model files for sending to the factory client device, identifying a plurality of candidate 3D model files stored in the model file library based on a measure of similarity between the extracted features and the tagged features for each of the candidate 3D model files;
sending the plurality of candidate 3D model files to the customer device to be displayed to the customer; and
receiving, from the customer device, a selection by the customer of one or more of the candidate 3D model files, wherein the selection of the one or more 3D model files for sending to the factory client is performed based on a measure of similarity between the modified extracted features and the tagged features for each of the 3D model files.

8. The non-transitory computer-readable storage medium of claim 6 , wherein the descriptive information does not comprise a 3D model file of the part.

9. The non-transitory computer-readable storage medium of claim 6 , wherein extracting one or more features from the descriptive informatio

Assignees

• Desprez Llc

Inventors

• Schwartz Jonathan
• Friefeld Max
• Ortlieb Oliver

Classifications

• Y02P90/02

  Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS] · CPC title

• G05B2219/49005

  Map 2-D pattern on 3-D · CPC title

• B33Y50/02

  for controlling or regulating additive manufacturing processes · CPC title

• B33Y50/00

  Data acquisition or data processing for additive manufacturing · CPC title

• G05B2219/49007

  Making, forming 3-D object, model, surface · CPC title