Printer from Computer Science and Artificial Intelligence Lab uses machine vision and 3-D scanning to self-correct and directly embed components.
3-D printing is great, assuming you're printing one material for one purpose, and that you’re fine with a few do-overs. But the technology is still far behind in reliably producing a variety of useful objects, with no assembly required, at a moderate cost.
In recent years, companies have been working to tackle some of these challenges with “multi-material” 3-D printers that can fabricate many different functional items. Such printers, however, have traditionally been limited to three materials at a time, can cost as much as $250,000 each, and still require a fair amount of human intervention.
But this week, researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) say that they’ve found a way to make a better, cheaper, more user-friendly printer. In a paper accepted at the SIGGRAPH computer-graphics conference, a CSAIL team presented a 3-D printer that can print an unprecedented 10 different materials at once by using 3-D-scanning techniques that save time, energy, and money.