US 11,809,162 B2
Methods and software for manufacturing a discrete object from an additively manufactured body of material including a precursor to a discrete object and a reference feature(s)
James L. Jacobs, Rye Beach, NH (US); and Arthur Richard Baker, Excelsior, MN (US)
Assigned to Protolabs, Inc., Maple Plain, MN (US)
Filed by Protolabs, Inc., Maple Plain, MN (US)
Filed on Mar. 19, 2021, as Appl. No. 17/206,313.
Application 17/206,313 is a continuation in part of application No. 16/454,166, filed on Jun. 27, 2019, granted, now 10,983,506.
Application 16/454,166 is a continuation in part of application No. 14/172,462, filed on Feb. 4, 2014, granted, now 10,373,183.
Claims priority of provisional application 61/891,453, filed on Oct. 16, 2013.
Prior Publication US 2021/0263500 A1, Aug. 26, 2021
Int. Cl. G06T 15/00 (2011.01); G05B 19/4155 (2006.01); G06N 20/00 (2019.01); B33Y 30/00 (2015.01); B33Y 50/02 (2015.01); G06F 30/10 (2020.01); G06F 113/10 (2020.01)
CPC G05B 19/4155 (2013.01) [B33Y 30/00 (2014.12); B33Y 50/02 (2014.12); G06N 20/00 (2019.01); G05B 2219/31368 (2013.01); G06F 30/10 (2020.01); G06F 2113/10 (2020.01)] 18 Claims
OG exemplary drawing
 
1. An automated manufacturing system for generating a graphical representation of a discrete object to be manufactured from an additively manufactured body of material, including a precursor to a discrete object and at least a reference feature, at a computing device, the automated manufacturing system is designed and configured to:
receive a graphical computer model of the at least a precursor to the discrete object and a graphical computer model of the discrete object;
identify at least a first feature in the graphical computer model of the discrete object, and recommended joining data;
selecting a correlated dataset containing a plurality of data entries wherein each dataset contains a geometric description of at least a stored critical-to-quality shape description and a correlated shape that requires manufacture to the first tolerance;
training a machine-learning model with the correlated dataset; and
automatedly determine, at the machine-learning model, that the at least a first feature includes at least a critical-to-quality feature in the graphical computer model of the discrete object, wherein automatedly determining at the machine-learning model comprises:
determining, at the machine-learning model, the geometric description of the at least a surface feature matches the geometric description of the correlated dataset as a function of comparing the geometric description of the at least a surface feature to the correlated dataset, wherein the machine-learning model is trained by the correlated dataset;
automatedly generate, at the machine-learning model, a graphical representation of the at least a reference feature on the graphical model of the at least a precursor to the discrete object as a function of the at least a critical-to-quality feature, at least a locating feature in a support, and the recommended joining data, wherein the at least a critical-to-quality feature corresponds to a shape that requires manufacture to a first tolerance based on the correlated dataset, and wherein the support comprises a substrate for deposition of layers in an additive process; and
automatedly determine a second feature as a non-critical to quality feature to be manufactured to a second tolerance which is lower than the first tolerance, wherein the non-critical to quality feature comprises surface finish or fit.