US 11,931,950 B2
Machine learning informed control systems for extrusion printing processes
Brian Howell, Livermore, CA (US); Brian Giera, Oakland, CA (US); Maxwell Murialdo, Westminster, CA (US); and Kyle Sullivan, Brentwood, CA (US)
Assigned to LAWRENCE LIVERMORE NATIONAL SECURITY, LLC, , CA (US)
Filed by Lawrence Livermore National Security, LLC, Livermore, CA (US)
Filed on Sep. 5, 2019, as Appl. No. 16/562,253.
Claims priority of provisional application 62/746,489, filed on Oct. 16, 2018.
Claims priority of provisional application 62/739,068, filed on Sep. 28, 2018.
Prior Publication US 2020/0101670 A1, Apr. 2, 2020
Int. Cl. G06N 3/02 (2006.01); B29C 64/106 (2017.01); B29C 64/393 (2017.01); B33Y 10/00 (2015.01); B33Y 50/02 (2015.01); G06F 18/24 (2023.01); G06N 3/008 (2023.01); G06N 20/00 (2019.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC B29C 64/106 (2017.08) [B29C 64/393 (2017.08); B33Y 10/00 (2014.12); B33Y 50/02 (2014.12); G06F 18/24 (2023.01); G06N 3/008 (2013.01); G06N 3/02 (2013.01); G06N 20/00 (2019.01); G06V 10/454 (2022.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A method performed by one or more computing systems for controlling a material extrusion device to extrude a filament of an ink to manufacture a product, the method comprising:
collecting one or more measurements from one or more sensors that monitor material extrusion processing of the material extrusion device, each measurement representing an internal state of the material extrusion processing during extrusion of the filament;
collecting an image of a portion of the filament while the filament is being extruded and while the portion of the filament is in motion and not in contact with any surface;
applying a classifier to the collected image of the filament as the filament is extruded, to generate an image-derived state characterizing the filament, the classifier being trained with training data that include images labeled with image-derived state data;
estimating a derived state of the material extrusion processing using a model based on the internal state and the image-derived state;
determining control parameters using the model to achieve a desired quality of the filament by minimizing a cost function based on the internal state, the image-derived state, the derived state, and constraints of the material extrusion device; and
providing the control parameters to a controller of the material extrusion device.