Grey-1 artificial intelligence, capable of autonomously detecting and correcting errors during the 3D printing process

A team of researchers from the University of Cambridge will includeEnder 3 Proconsumer grade, including3DThe printer switches toAIMotivated “self-learning”3Dprinter, and earlier this month atRedditA fascinating article was published about thiswidelyfollowed, and the results were shared in several research articles.

Today, a startup formed by this research group on computer-aided manufacturingMattaThe development of this technology continues to make it aGray-1: A device that uses machine vision to autonomously correct flow rate, feed speed, hot end temperature andZaxis offsetAI“Make a co-pilot.” It works with a variety of printers and materials, includingTPUnylon filled with carbon fiber, silicone, etc.Material。

Matta“The system is basically trained by showing the model3DThe different reasons why printers may malfunction and how to resolve these errors. With enough examples and the right algorithms, the system can learn to recognize at any time whether an error has occurred and then take appropriate action.plancorrect.»

Brian andMattaThe team is currently working to integrate its technology into the slicing process. If successful,Gray-1Performance during the printing process will be monitored and this information will then be used to automatically adjust cutting parameters for future parts.

Although the project is focused on commercial use, the team also plans to release aOctoPrintAndRocksThe plugin isRaspberry Pirun onGray-1version, which can announce real user-friendliness3DPrinting experience.

an aspectpublicofGray-1Seems like it’s a matter of ease of use3DA giant step forward in the field of printers.bigBrands already seem to be moving towards this goal, asPrusa SearchofNexttruderPromises a “perfect first layer”, but we haven’t seen anything like that in standalone products.

at present,MattaNow available on the siteGray-1interactive demonstration. Additionally, Brian recentlyRedditDuring a Q&A session on , he provided a link to the team’s research paper, revealing moreMattaA team effort and definitely worth the read.

Source: all 3dp