MIT Invents A Flowing River Of 3-D Pixels That Lets Objects Assemble Themselves

The Tangible Media Group’s latest innovation could revolutionize the assembly line. Here’s how.

When MIT’s Tangible Media Group first unveiled its shapeshifting display, the inFORM, one thing the team pointed out was that it had a lot of possibilities for the manufacturing and industrial sectors. With Kinetic Blocks, a follow-up of sorts to the inFORM, the Tangible Media Group chose to explore this potential, showing how the conveyor-belt assembly lines of today could be replaced by flowing rivers of 3-D pixels.


Like the inFORM, Kinetic Blocks is a flatbed “shape display” made up of computer-controlled pins, with a Microsoft Kinect as an overhanging eye. What has been supercharged here is the ability and granularity with which those pins can manipulate objects. In their video demonstration, the Tangible Media Group shows how the Kinetic Blocks platform can be used to stack, rotate, twist, and move blocks, without any human intervention. It can even construct preprogrammed structures.

It’s true that the original inFORM could manipulate objects to some extent. But the execution was clumsy, and it required a human operator to manipulate an object on top of the display. Not only is Kinetic Blocks fully automated and programmable, but it’s much, much more speedy and accurate at manipulating the objects you place within its field. You can also use Kinetic Blocks to record the motions of a human operator assembling something. The program will then analyze the human’s motions, and learn how to recreate the assembly from scratch.

As with the inFORM before it, you have to use your imagination a little to see the application of shape displays, but once you do, you can’t unsee it. Right now, assembly lines require expensive human or robot operators to assemble a product. Something like Kinetic Blocks could transform the conveyor belt of an assembly line into a robot operator. Imagine automobiles or iPhones that seemingly assemble themselves as they move down the factory line. Kinetic Blocks isn’t capable of making that happen yet, mostly because it doesn’t have enough pixels to finely manipulate small objects, but the concept proves it’s possible: All we need is for shape displays to go Retina.

[via Design Boom]