A Stabilizing Robotic Tail for Floating Astronauts | Hacker News Reader

A Stabilizing Robotic Tail for Floating Astronauts

core77.com

102 pointssizzle43 comments

in the anime Mobile Suit Gundam it's an established concept that the suits used 'AMBAC' to coordinate in space[0], which is really just a computer coordinated method of flailing the limbs on the robot to get inertia and mass to do its' thing to move the robot without thrusters, similar to a mass wheel but without the needed additional weight of a flywheel.

I always sort of wished that they had animated the little micromovements into the suits to show this feature more readily ,but production costs and weird looking movement generally prevented AMBAC from ever being demonstrated visually.

This concept reminds me a lot of that.

[0]: https://gundam.fandom.com/wiki/Active_Mass_Balance_Auto-Cont...

Reminds me of Ardent, from Alice Grove: https://www.questionablecontent.net/alice1.html

Thank you for linking that, it was a very enjoyable read.

Jeff Jacques is best known for QC, but he’s done other stuff, over the years.

I really enjoyed Alice Grove. I think I remember reading, somewhere, that he’s working on something similar, these days.

A lot more impressive than I expected, with a combination of Kevlar/Carbon black filled UHMWPE/PEEK and radiation shielding it might actually be able to do the work asked of it.

A cool device. Seems like its completely autonomous. Seeing as we used to have tails in our evolutionary past, I wonder if this could be fitted like a myoelectric prosthesis around the coccyx or something and function like a biological tail.

Attempting to replicate a tail seems tangential to the purpose of the device. The human coccyx is a poor location to anchor anything given how the vertebrae are not equipped to handle force, and there is effectively zero musculature to support it. The harness on the back appears much more practical.

Let's assume that for some reason attaching sensors/stimulatirs to the coccyx area lets us control tail like robots. Then the sensors could be attached in that area, and the actual tail could be mounted anywhere else, perhaps even remote to the astronaut's body.

I'm not sure that I buy that first assumption though. Measuring signals might be possible, and it might be possible to learn how to move the tail based on signals produced during conscious attempt to move the tail. But I don't see how it would be possible to send signals back to the brain.

That's an assumption that isn't rooted in anatomy. You need nerves to measure signals on. Look up the comparative anatomy of the human cauda equina vs that of a dog. Where you're thinking of plugging into the body, in a human you'd be left with the filum terminale, which is going to sorely disappoint if you are looking for nerve signals.

To get any meaningful spinal communication in this hypothetical scenario, you'd be looking higher up in the sacral region. Unless this hypothetical also includes some magic star trek technology, you'd be invasively implanting electrodes there, which sounds like a fantastic way to trigger fecal incontinence and chronic pain.

I get the fun idea of returning humans to our primate origins, but modern human anatomy has left us effectively nothing to work with. It's an anatomical dead end.

Right, thanks for the proper information. I guess you need to go pretty far into our past to get to a tailed ancestor, quick google says its 25-30myam, pre-hominid monkeys. Back scratcher manufacturers aren't going out of business anytime soon.

Completely naïve guess, but maybe it'd make sense to tap into the hip muscles to hint the controller system what you want the tail to do. In zero g, a slight rotation in the pelvis won't cause too much movement, but I think it would feel fairly natural to trigger a tail swing with a hip motion.

I think the tail attached to the back might be physically awkward or impractical. After all, there's probably a good reason that all animals with prehensile tails have it attached to the pelvis. But then again, it may just be an artifact of it being an extended spine.

Unironically, I think this is a fantastic idea that might actually catch on if/when humans start spending extended time in zero g environments. Tails do have a lot of advantages for navigating and manipulating 3D space.

Unfortunately, I don't think we'll be able to actually solve the problems with this design until we're actually in space regularly.

Then it could be prehensile, with space tool attachments...

> Seems like its completely autonomous.

The final / functional version should be; as it stands though, it's a prototype / concept. It'll need a lot of sensors and fine-tuning to work as intended though.

Cool. The demos see. To be mostly about grabbing.

Would be great if it also helps to maintain flight direction, like a squirrel tail.

But probably difficult to test in weightless conditions on a student project budget.

The wings are doing a lot more work than the tail in that situation. For quick rotations in 3D, gyroscopic forces would probably be better to utilize than a tail

Doesn't need to be in zero-G, just dangling a dummy off a harness should be enough. For attitude control I'd imagine gyros would be much more pragmatic, like a smaller scale 3D version of https://www.roborigger.com.au/

Furries been making tail cores like this for years

Furries! In! Space!

Very cool! Though I'm not convinced this isn't just some covert furry sneaking his gear into the mainstream.

Dr. Otto Gunther Octavius, I presume?

I wonder if this is inspired by the Ousters from Dan Simmons' Hyperion, who're described as using a prosthetic tail to move around in zero-g environments.

For moving in zero-G, I wonder if a Batman-style grapple-gun might be useful. Make the grapple slow and maneuverable so that once you fire it, it will always grab on to a loop or bulkhead or whatever.

But for floating in place, that tail feels like the right solution. Although if you did have a magical automatic grapple gun, maybe using 3 of them at once would also keep you in place very well. (although the cables might get annoying!)

I love when engineering outpaces SF.

My first thought on seeing this was, I can’t believe the Belters never thought of this in The Expanse.

The string control is ingenious! Way more appropriate than a bunch of stepper motors.

I remember seeing inflatable designs too.

Perhaps your assumption that stepper motors are the first solution to any motion problem speaks more about your lack of resourcefulness than the ingenuity of another solution.

It was a comment on the overengineering that is rampant in tech nowadays. What's with your attitude?

Using individual actuators is usually the simplest solution to a problem.

Reminds me of the robotic hand from big bang theory. I could use a tail. This seems like one of those things that might catch on with normies. They should do a product launch.

Ugh, another site that fucks with scrolling.

I don't know what's going on with that site but it partially loads and a few seconds later I get a warning that JS is slowing Firefox down. Clicking stop does nothing, hitting the back arrow does nothing, etc. I had to close the tab completely.

Here's an alternative link: https://www.hackster.io/news/mit-s-robotic-tail-could-aid-cr...

...and the MIT media lab page about it: https://www.media.mit.edu/articles/mit-s-robotic-tail-could-...

It's a really cool idea that I don't think I've seen in the scifi I've watched/read. But given this is a Media Lab project I expect the claims and capabilities to be wildly overstated and the project to go nowhere.

It's not the Media Lab project. It is a project by an RCA student in London https://2022.rca.ac.uk/students/cheng-chang/

All I hear is "tails are useful in space". Monkeys have tails. Monkeys descended from space. What if it was aliens?!

Wondering why this needs to be a tail - this could be Doc Ock's extra arms.

Ok so I have to be the one to do it: "Return to monke"

one day we'll all be space monkeys!

Marsupilami!

If you click “Manage Options” on the data collection popup, the switches toggle between two slightly different shades of grey, which is mildly infuriating.

> there is going to be a whole generation of humans living their lives in zero-gravity

No there isn't, although you might have some cosmonauts doing a few years of it, it's now a long term liveable condition.

Yeah, right now we can't even get a couple years without serious health effects in adults.

I assumed it'd be about automated angular momentum rather than gripping things.

Like how animals in real life use tails.

Just rather than on earth it's balance, it'd be consistent direction.

I had the same expectations. Would be a super difficult control problem I think, with stuff like the end of the tails motion range, collisions, lack of ability to predict the motions of actual limbs and so on. Perhaps a 2D model could be a useful approach to this problem, with super low friction swivel chairs too high for the feet to interact with the ground?

Might even be an interesting playground for Neuralink. Robotic tails with a diffuse control from the brain, now that would put the "furries" comments elsewhere in the thread on the next level!

I've seen a few EEG based prosthetic controls in pre-pandemic cons. I think one was a DIY hacker project to control a tail, another was a commercial thing just for ears: https://en.wikipedia.org/wiki/Neurowear

Reply on news.ycombinator.com