Systems, methods, and media for artificial intelligence process control in additive manufacturing
Systems, methods, and media for additive manufacturing are provided. In some embodiments, an additive manufacturing system comprises: a hardware processor that is configured to: receive a captured image; apply a trained failure classifier to a low-resolution version of the captured image; determine that a non-recoverable failure is not present in the printed layer of the object; generate a cropped version of the low-resolution version of the captured image; apply a trained binary error classifier to the cropped version of the low-resolution version of the captured image; determine that an error is present in the printed layer of the object; apply a trained extrusion classifier to the captured image, wherein the trained extrusion classifier generates an extrusion quality score; and adjust a value of a parameter of the print head based on the extrusion quality score to print a subsequent layer of the printed object.
1. An additive manufacturing system, comprising:
a print head that is configured to print an object in a layer by layer manner;
an image sensor configured to capture an image of a printed layer of the object; and
at least one hardware processor configured to:
receive the captured image;
sample the captured image to generate a low-resolution version of the captured image;
apply a trained failure classifier to the low-resolution version of the captured image;
determine, based on a classification obtained from the failure classifier, that a non-recoverable failure is not present in the printed layer of the object;
in response to determining that a non-recoverable failure is not present in the printed layer, generate a cropped version of the low-resolution version of the captured image;
apply a trained binary error classifier to the cropped version of the low-resolution version of the captured image;
determine, based on a classification obtained from the binary error classifier, that an error is present in the printed layer of the object;
in response to determining that the error is present in the printed layer of the object, apply a trained extrusion classifier to the captured image, wherein the trained extrusion classifier generates an extrusion quality score indicating a quality of extrusion of material in the printed layer of the object; and
adjust a value of a parameter of the print head based on the extrusion quality score to print a subsequent layer of the printed object.
2. The additive manufacturing system of claim 1 , wherein the cropped version of the low-resolution version of the captured image is generated by identifying a region of interest that includes a mesh pattern representative of the extrusion quality score.
3. The additive manufacturing system of claim 1 , wherein the hardware processor is further configured to:
receive a second captured image of the subsequent layer of the printed object;
sample the second captured image to generate a low-resolution version of the second captured image;
apply the trained failure classifier to the low-resolution version of the second captured image;
determine, based on a second classification obtained from the failure classifier, that a non-recoverable failure is present in the subsequent layer of the object; and
in response to determining that a non-recoverable failure is present in the subsequent layer, terminate printing of the object.
4. The additive manufacturing system of claim 1 , wherein the failure classifier comprises a Convolutional Neural Network (CNN).
5. The additive manufacturing system of claim 1 , wherein the extrusion classifier comprises a plurality of CNNs.
6. The additive manufacturing system of claim 1 , wherein the parameter of the print head is an extrusion volume.
7. A method for additive manufacturing system, comprising:
receiving the captured image;
sampling the captured image to generate a low-resolution version of the captured image;
applying a trained failure classifier to the low-resolution version of the captured image;
determining, based on a classification obtained from the failure classifier, that a non-recoverable failure is not present in the printed layer of the object;
in response to determining that a non-recoverable failure is not present in the printed layer, generating a cropped version of the low-resolution version of the captured image;
applying a trained binary error classifier to the cropped version of the low-resolution version of the captured image;
determining, based on a classification obtained from the binary error classifier, that an error is present in the printed layer of the object;
in response to determining that the error is present in the printed layer of the object, applying a trained extrusion classifier to the captured image, wherein the trained extrusion classifier generates an extrusion quality score indicating a quality of extrusion of material in the printed layer of the object; and
adjusting a value of a parameter of the print head based on the extrusion quality score to print a subsequent layer of the printed object.
8. The method of claim 7 , wherein the cropped version of the low-resolution version of the captured image is generated by identifying a region of interest that includes a mesh pattern representative of the extrusion quality score.
9. The method of claim 7 , further comprising:
receiving a second captured image of the subsequent layer of the printed object;
sampling the second captured image to generate a low-resolution version of the second captured image;
applying the trained failure classifier to the low-resolution version of the second captured image;
determining, based on a second classification obtained from the failure classifier, that a non-recoverable failure is present in the subsequent layer of the object; and
in response to determining that a non-recoverable failure is present in the subsequent layer, terminating printing of the object.
10. The method of claim 7 , wherein the failure classifier comprises a Convolutional Neural Network (CNN).
11. The method of claim 7 , wherein the extrusion classifier comprises a plurality of CNNs.
12. The method of claim 7 , wherein the parameter of the print head is an extrusion volume.
13. A non-transitory computer-readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for additive manufacturing, the method comprising:
receiving the captured image;
sampling the captured image to generate a low-resolution version of the captured image;
applying a trained failure classifier to the low-resolution version of the captured image;
determining, based on a classification obtained from the failure classifier, that a non-recoverable failure is not present in the printed layer of the object;
in response to determining that a non-recoverable failure is not present in the printed layer, generating a cropped version of the low-resolution version of the captured image;
applying a trained binary error classifier to the cropped version of the low-resolution version of the captured image;
determining, based on a classification obtained from the binary error classifier, that an error is present in the printed layer of the object;
in response to determining that the error is present in the printed layer of the object, applying a trained extrusion classifier to the captured image, wherein the trained extrusion classifier generates an extrusion quality score indicating a quality of extrusion of material in the printed layer of the object; and
adjusting a value of a parameter of the print head based on the extrusion quality score to print a subsequent layer of the printed object.
14. The non-transitory computer-readable medium of claim 13 , wherein the cropped version of the low-resolution version of the captured image is generated by identifying a region of interest that includes a mesh pattern representative of the extrusion quality score.
15. The non-transitory computer-readable medium of claim 13 , wherein the method further comprises:
receiving a second captured image of the subsequent layer of the printed object;
sampling the second captured image to generate a low-resolution version of the second captured image;
applying the trained failure classifier to the low-resolution version of the second captured image;
determining, based on a second classification obtained from the failure classifier, that a non-recoverable failure is present in the subsequent layer of the object; and
in response to determining that a non-recoverable failure is present in the subsequent layer, terminating printing of the object.
16. The non-transitory computer-readable medium of claim 13 , wherein the failure classifier comprises a Convolutional Neural Network (CNN).
17. The non-transitory computer-readable medium of claim 13 , wherein the extrusion classifier comprises a plurality of CNNs.
18. The non-transitory computer-readable medium of claim 13 , wherein the parameter of the print head is an extrusion volume.