"Fab By Example" Data-Driven Manufacturing Method for DIY Designs at Home

Even as 3D printing is poised to help democratize manufacturing, it's often overlooked that many 3D-printed items are far too complicated for users to digitally design. Now, researchers at MIT's Computer Science and Artificial Intelligence Lab (CSAIL) have developed 'Fab By Example,' the first data-driven method to help people design products, with a growing database of templates that allow users to customize thousands of complex items, such as cabinets, jungle gyms, and go-carts. "When we design things on a computer, the question arises of how to manufacture them in the real world with the necessary physical parts - wood, glass, screws, hinges, bolts, and all," says project lead Adriana Schulz. 'For casual users, creating such a detailed model is not just time-consuming, but it's actually more or less impossible unless you know something about mechanical engineering.' Where previous do-it-yourself design databases have required an advanced degree, or at least expertise in computer-assisted design (CAD) software, the team says that now even someone with simple computer skills can make a own customizable item.

Transcript

00:00:00 Real-world objects are fairly complicated. Often, their virtual representations are greatly simplified. When we compare this table model from Google Warehouse to its real-world counterpart, we immediately notice that many details are missing. These details are what make designing fabricable objects very difficult for casual users. In this work, we present a modeling tool

00:00:25 that leverages information from a collection of fabricable designs, in order to enable casual users to model and build real-world objects. To allow structure-preserving manipulations, we automatically convert the designs in our collection to hierarchical parameterized templates. Our template model takes into account global relationships, such as symmetries and articulation consistencies. It is also not restricted to fixed topologies, allowing

00:00:49 irregular patterns to be rescaled. Notice how the number and position of connectors changes, as the template parameters vary. Our modeling tool is based on template manipulation and composition. We allow users to explore the space of template variations using different levels of the template hierarchy. During manipulation, active elements of the hierarchy

00:01:12 are shown in full color, while other elements are semitransparent. Once a part is positioned, the user invokes a method that brings in connecting elements to the design, by exploring the template database. Notice how for our a wooden tabletop, corner brackets are automatically brought in to connect the top to the legs. While, for a glass stop, a whole support structure for the glass is automatically added.

00:01:34 We incorporate in our system a stability analysis evaluation that warns the user if the design model is unstable. We illustrate a full modeling session. For composition, we first assist the users by snapping drag components onto the working model. The user can continue to reposition and rescale the model, until she is satisfied with the snap configuration. The output of our system is a comprehensive bill

00:02:01 of materials that can be directly used for fabrication. Here are some additional results. One of the main advantages of using a data-driven approach is its generality. Notice how the same method that creates tables and cabinets can also be used on go-karts.