MIT inventor creates cube robots with no external moving parts | Hacker News Reader

MIT inventor creates cube robots with no external moving parts

kurzweilai.net

170 pointsLibertatea44 comments

-"The MIT researchers are currently building an army of 100 cubes"

Hardly "unleashes hundreds of swarmbots", more like unleashes eight of them, and plans to release more .. Quite a misleading article title.

edit: I'm not complaining about the HN title, which is just taken from the article. I'm pointing out that the actual article title, which states that hundreds of these bots have been unleashed (lie, it's about eight) and that they self assemble (lie, they require direct instruction) is misleading...

If you follow the official mit.edu link, you can find that statement in the last paragrah.

http://web.mit.edu/newsoffice/2013/simple-scheme-for-self-as...

Agreed. Terms like army and unleashed as well as the Terminator 2 reference in the article attempt to jazz up a piece of work which is interesting enough in it's own right.

It encourages you to imagine what these things can be good for, therefore get excited about even just this small progress.

It encourages me to be disappointed when I actually read the article, which is not exactly the best state to put me in if you want me to appreciate your article.

From the video at the bottom "We found out by accident that they were able to jump, we weren't intending for them to do that..."

Maybe the Terminator comparisons were all too prescient...

"We also found out that they were able to extend spikes from their bodies. We don't know how they--urrrrrrk!"

Mistook self-assembling for self-replicating and became unduly alarmed.

Same. I really, really hope no researcher/terrorist/researcher-terrorist releases some kind of grey goo in the next hundred years. That sounds like a terrible way to die.

We already have something similar covering the planet -- except most of it is green. Some of it is gray though.

Jesus, this is blatant blogspam. The OP is much more interesting and has more detail: http://web.mit.edu/newsoffice/2013/simple-scheme-for-self-as...

Tiny bit too breathless and too much hype.

I've seen children's toys that used internal momentum via a counterweight back in the 80s.

Well, I don't know about terminators or replicators but they could do a self-solving rubik's cube once they'll have built 26 of these blocks.

What does everyone think about energy use and storage? Are magnetic devices energy efficient enough for this to work some day?

Maybe they could attach them to an external power supply?

Seems like it could be a constraint that needs to be solved for this to be useful

Energy use seemed like a significant issue - how many flywheel spinups do you get out of a battery?

They could include a charging circuit that routes power through the assembled blocks. Each unit would still require it's own independent power for disconnected motion, but they could recharge when connected.

These would be perfect office desk toys. Right up until they combine and take over the office of course.

Esp. their design reminds me rather of Stargate's good old Replicators than Terminator ;)

The replicators were worse than the terminators....

I think these robots are missing a few things that would make it actually useful. They need a latching mechanism to increase strength when they are in position. And they need a manipulator that can bend while locked onto two other cubes.

When I see these I think of a flexible exoskeleton or simulated bones. They are missing what would be the cartilage and muscle of a moving being. But who knows this could be a great step in that direction.

I was thinking the same thing until I saw how they rotated along the paired edge. They don't need "cartilage and muscle", they're machinery, they need a hinge and the ability to apply force to move that hinge. With a stronger magnetic point along they edge, the swing of a hinge can be accomplished, though it will take some very fine tuning of the actuator and braking to accomplish something like walking, but I suspect it can be done.

A step, indeed; this is research, it need not be actually useful. The things you mention, I can see as "left as an exercise for the reader".

I am more worried about power efficiency. Corrections welcome, but I think accelerating and (rapidly, in many cases) decelerating a flywheel can't be that efficient. Also, I don't see how those flywheels could efficiently move a large assembly of these things. Let's say you have a 5x5x5 cube: how would they work together to move that? Fall apart and reassemble?

I actually think that spinning up a momentum wheel is fairly low energy, but it's been a long time since we did that.

Like I said, if they locked together you could get movement from a huge cube. But I think the key in these guys is to build dynamic rigid bodies and have "joints" that move around to connect them and help them actuate. That'd be awesome.

They could also snake along the ground doing a chain where one goes from the back to the front, then the next and so on. Anyway, pretty cool idea to start from. I can't wait to see if they iterate this and add cameras and a main controller or anything like that!

I'm pretty sure I have read of a similar self assembling cubes concept in a science magazine when I was a kid (i.e. 20 years ago). The concept was only rendered in 3D (which was a piece of cake even then for it was only cubes). Pretty neat to see that come to reality.

Scaled down to 1/1000th of it's current size and that would be an interesting way to perform complex operations, like brain surgery. No more cutting in a straight line, instead you can move around objects. Though by that time we'd probably have little self-propelled nano-bots doing this work more efficiently.

Dynamic stabilization would be interesting. This could "fill" in an area, say the sudden damage to a building, perhaps even a space station or spaceship.

Movement could be another feature. This could serve as wheels to move over complex terrain (up mountains, through the bottom of the ocean). They could go from fins, to wheels, to little hands or claws, etc...

How would the robot be able to maintain coherence at such a small size?

This technology has improved nicely over the years. For comparison, here are some similar swarm bots from 2005: http://www.youtube.com/watch?v=seGqyO32pv4

The first thing I thought about when I watched the video is T-1000.

>The experts said it couldn’t be done. But research scientist John Romanishin of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) has created M-Blocks — cube robots with no external moving parts.

I doubt that the experts were thinking about 1cm blocks when they passed their judgment.

I have to admit I was extremely disappointed with the video after reading the title. 8 cubes with motors that can move via radio commands? Astonishing...

Agreed, but give them credit for building something new that arguably fits under "basic science" and might have unexpected applications in the future. For instance, I could imagine a system like this for reconfiguring parts of a satellite after it has been launched into space.

Perry the platypus needs nip this on the bud

As one with kids that love Phineas and Ferb: Thank you.

seem to be an interesting thing. Yes the article is of somewhat lesser quality than the MIT own, but Thank you for post it never-the-less. I'd hope they will try to open a bit their project, so technically like minded people will be able to contribute. :)

I bet the results would be fascinating if we could apply some cellular automata rules to these.

I, for one, welcome our new Grey Goo [1] Overlords ;-)

Let's just hope those don't develop self-preservation instinct anytime near soon.

[1] http://en.wikipedia.org/wiki/Grey_goo

Different kind of self-assembling.

Michael Crichton's book Prey might not be too far off.

>We want hundreds of cubes, scattered randomly across the floor, to be able to identify each other, coalesce, and autonomously transform into a chair, or a ladder, or a desk, on demand

Sounds interesting. But when will I get a robot to cook my meals, clean my apartment, and do my laundry?

And after that, when will we be doing theirs?

Microwave oven, vacuum cleaner (Roomba), and washing machine :)

It's very cool and very clever from a hacking and engineering standpoint but I don't see what they're aiming for.

It seems they want to make groups of cubes that can autonomously reconfigure themselves, but what will they do once they get that? What task could they achieve?

I guess it could make a fun toy but I can't think of anything else.

"But the researchers believe that a more refined version of their system could prove useful even at something like its current scale. Armies of mobile cubes could temporarily repair bridges or buildings during emergencies, or raise and reconfigure scaffolding for building projects. They could assemble into different types of furniture or heavy equipment as needed. And they could swarm into environments hostile or inaccessible to humans, diagnose problems, and reorganize themselves to provide solutions." http://web.mit.edu/newsoffice/2013/simple-scheme-for-self-as...

Self assembling furniture

Imagine just buying a crate and unloading it in your living room.

Obviously they're far away from something that could create visibly pleasing furniture.. But this is just the first step

Science and technology are incremental. Someone needs to take the first step before the second can be taken.

Reply on news.ycombinator.com