Mobile 3D Printer Can Print Directly on Your Floor | Hacker News Reader

Mobile 3D Printer Can Print Directly on Your Floor

spectrum.ieee.org

49 pointsrbanffy24 comments

I view this as closer to conceptual performance art than a practical solution to an engineering problem.

TFA article talks about how the robot “feels” and the “relationship” between the printer and the environment. That’s perfectly valid but it is artist’s statement stuff.

What I didn’t see is any discussion about taking the fixed coordinate system defined by a printer’s volume and translating that into the larger space, which would presumably need to be mapped to 0.1 mm precision. Or at least any previously printed objects would need to act as registration points at that level of precision.

I'm definitely giving this one a chance. It's version 1. I'm convinced that many good things will evolve from this.

I can already print human scale objects domestically on a belt printer. But only on a single axis. Something like this would have the advantage of being able to print at human scale on all axes. But you’d need a human sized printer on at least one axis. At which point, you have to question what you’re doing.

It would be a fun project to figure out how to optimize the support material necessary for this printer to raise itself up and move itself around a print that is far larger than itself.

Sort of like scaffolding for commercial construction.

you mean for the printer to print its own scaffolding?

Any scenario where the printer itself moves but continues to print an in-progress object requires the equivalent of auto-bed-leveling. But rather than checking the location of a fixed bed with known parameters, it would probably be checking the unfinished face of the printed object. Sounds challenging...

Yes, exactly, print it's own scaffolding. Sounds very challenging.

Imagine this robot printing a human sized David statue and the necessary scaffolding to raise itself to every position necessary to do so.

The next level would be to design the scaffolding in such a way that it's easy for the robot to disassemble and either directly reprocess into filament or to store for later reprocessing.

The next level would be coordinating a swarm of these robots.

> What I didn’t see is any discussion about taking the fixed coordinate system defined by a printer’s volume and translating that into the larger space,

That's because the creator didn't make it do that. I was looking for this as well but a careful reading shows that the print volume is the same as a normal Prusa Mini.

It's a Prusa Mini on a Roomba, that's it. It can drive to a spot and then print there normally, nothing else. What you are describing would be a far more interesting project, in my opinion.

The engineering problem is 3D printer build volume limits.

For those who aren't familiar, poor first layer adhesion to the print surface is one of the most common causes of print failure in 3D printing. Typically you want a very clean build surface made of an appropriate material, and sometimes covered in glue stick or hairspray.

I can imagine they went through a lot of false starts with this.

I can't imagine this working that well on a cold floor either.

It can always spray some adhesive before it starts printing.

Could the vacuum clean the floor first?

At a minimum you would want to not only vacuum the floor, but mop it with isopropyl alcohol. However household floors aren't even and level enough and aren't made of the right material.

People reading this may also like hexapod from (my goodness) 16+ years ago, which was a little robot cnc machine

https://www.youtube.com/watch?v=7hEXwyJ2B78

https://www.youtube.com/watch?v=quN37YskoaM

“MobiPrint, designed by Daniel Campos Zamora at the University of Washington, consists of a modified off-the-shelf 3D printer atop a home vacuum robot.”

Love the creativity / repurposing.

Is the design/parts/source available?

Can it print furniture?

"Currently, MobiPrint can only “park and print.” The robot base cannot move during printing to make large objects, like a mobility ramp. Printing designs larger than the robot is one of Campos Zamora’s goals in the future. "

Well, you could put a multi-axis robot arm with sufficient reach on top of the base, a counterweight on one arm (or more) and some laser interferometry to position the head very precisely in relation to the room, floor, walls, and printed article. Would be interesting to see it stretching the printer arm while counterbalancing it with the other arms like a ballerina.

Ok so in a nutshell they took a turtlebot and stuck a 3d printer on top of it. Crazy what qualifies as an IEEE article these days.

Wasn't there a working implementation of cooperative holonomic robots a few years back that each had a print head and they could print arbitrarily large parts by driving around instead of using a gantry for the xy axis? I'm trying to find the article but Google just swarms me with a quadrillion results of how to 3D print a mecanum robot so there's that.

> Printing designs larger than the robot is one of Campos Zamora’s goals in the future.

https://en.wikipedia.org/wiki/Turtle_graphics

Not exactly new: some domestic pets are also mobile 3D printers capable of printing directly on your floor.

So if we mount this to a Spot or a Unitree Go... we have the pinnacle of robotic 3D-printed poop.

I hear the interplay of this "feature" with early robotic floor vacuums could create impromptu artworks.

Reply on news.ycombinator.com