Although scientists and engineers are continually inventing new materials with special qualities that may be utilized for 3D printing, doing so can be a challenging and expensive task.
In order to find the optimal settings that consistently produce a new material’s best print quality, an expert operator frequently has to do manual trial-and-error experiments, sometimes creating several prints. Printing speed and the amount of material the printer deposits are some of these parameters.
Artificial intelligence has now been employed by MIT researchers to simplify this process. They created a machine-learning system that watches the production process using computer vision and corrects handling errors in real-time.
They put the controller to a real 3D printer after using simulations to train a neural network how to change the printing parameters to reduce error. Compared to other 3D printing controllers, their technology produced items more precisely.
The procedure of printing thousands of millions of actual items to teach the neural network is avoided by the work. Additionally, it could make it simpler for engineers to integrate novel materials into their prints, enabling them to create items with distinctive chemical or electrical properties. If unexpected changes occur in the environment or the material being printed, it could also make it easier for technicians to make quick modifications to the printing process.
Wojciech Matusik, MIT’s Electrical Engineering And Computer Science Professor, said “This project is really the first demonstration of building a manufacturing system that uses machine learning to learn a complex control policy,” Matusik is a leader in the Computational Design and Fabrication Group (CDFG) which is in the Computer Science and Artificial Intelligence Laboratory (CSAIL).
“If you have manufacturing machines that are more intelligent, they can adapt to the changing environment in the workplace in real-time, to improve the yields or the accuracy of the system. You can squeeze more out of the machine,” he added.
Mike Foshey, CDFG’s Mechanical Engineer and Project Manager, and Michal Piovarci, a postdoc at the Institute of Science and Technology in Austria, are the research’s co-lead authors. Jie Xu, a Graduate student of Electrical Engineering and Computer Science, and Timothy Erps, CDFG’s former technical associate, are co-authors at MIT.
Send us a message!
What do you think about this post? Share your thoughts, comments, or suggestions. Feel free to ask questions or for more details. Just click the button below.