Detecting 3D Printing Fails
Although 3D Printing has a promising future, a common problem is that it is easy to mess up on a project, whether it's because of a malfunction in the 3D printer or an incorrect design. This results in wasted filament, and forces the designer to start over again, which can be frustrating since 3D printing something can take days.
Our app, Christine's 3D Printing, tackles all of these problems. Through artificial intelligence and machine learning, it has read thousands of images that are labeled as either 3D printing or 3D printing fail. The program is connected to a Raspberry Pi and camera, which can be set up in front of the 3D printer, and every minute, it takes a picture of the 3D printer to ensure that the process is running smoothly. The pictures are uploaded to our app, and the user will be alerted if the camera senses that the 3D printer is failing to work properly.
A program ran through thousands of 3D printing pictures and learned if they were fails or not.
We wrote a program in Visual Studio Code and transferred it to Flutter, an app development platform, to envision what the app would look like.
After setting up the Raspberry Pi and a camera, we coded to display the time and upload the pictures to a realtime database.
We set the camera in front of the 3D printer to test and detect live printing, touched up a few things like the camera focus.
These videos provides more information about how the app works when a 3D printer is printing live.