Hacker News new | past | comments | ask | show | jobs | submit login
Repurposing the cadaver of a spider to create a pneumatically actuated gripper (ieee.org)
153 points by latchkey on July 30, 2022 | hide | past | favorite | 108 comments



The itsy bitsy spider crawled up the pneumatic pipe. Along came the scientist and ripped its insides out. Now the itsy bitsy spider is an actuated gripper. La dee dee da da listen to teacher or your corpse will be a puppet.


Little Dr Muffet sat on her tuffet, eating her curds and whey. Along came the spider who sat down beside her, but she was not scared away.

Then came some thinking, necrotic controlling, hydrollic and terrifying beyond belief. Freezing the spider, and some air puffs beside her, she lifted some tinsel and hay.


Very tangentially related, though with AI and aliens, if you haven't read Blindsight by Peter Watts, you owe it to yourself to check it out.

https://en.wikipedia.org/wiki/Blindsight_(Watts_novel)


Much agreed, along with Greg Egan, Peter Watts writes stories that are probably perfectly suited towards the interests of HN'ers. Both are "hard sci-fi". Blindsight even has an appendix that lists sources and explanations for how concepts from the book tie back to things that can be seen in the real world.


Also check out Children of Time and it's sequel by Adrian Tchaikovsky.


Came here to plug this ... great book. (The sequel ... not so)


I enjoyed the second book except towards the end.


It definitely was not bad ... but compared to the first it was a bit off.


I read Blindsight based on a recommendation here.

Honestly… I think it’s a great book, but one that I would have to read multiple times to fully understand.


Yea, can confirm it benefits from re-reading. Without spoilers, there was one part in particular that was confusing on the first read.


Please spoil it, I've read the book multiple times and I'm looking to learn even more :)


For me, it was a bit too heavy on the science and too light on the rest.

The vampire stuff was funny, but felt a bit half assed.


I would argue that the sequel, Echopraxia, is very directly related. Even just the definition of the title.


Leaving the morbid details aside, it is remarkable how many of our industrial innovations are inspired by existing biology.

After all, the biosphere is an immense laboratory with experiments running nonstop for millions of years, and bad designs weeded out ruthlessly by evolutionary bottlenecks.


I wonder how many years were lost because inventors tried to replicate bird flight.

Wheels do not seem to be very widespread in the nature either.


Which ones do you have in mind? Making the argument the other way round seems a lot easier, for instance, wheels and axles weren't inspired by existing biology.


Freely rotating bearing surfaces don't really offer much compatibility with multicellular life. For nutrient/waste transfer or nervous systems to work between an outer body and an axle-like thing would be a big evolutionary step! It's fun to ponder how it might work and whether it could work robustly. We have the same kind of obstacles with machines featuring freely rotating assemblies: tire pressure sensors have to wirelessly communicate. Steering wheels, on the other hand, support wired connectivity thanks to a flexible connection and limited range of motion similar to joints in animals.


So basically it means it can only build by mutualism. A wheel organism and an axle organism, both feeding independently and somewhat connecting. Maybe male/female like anglefish


It'd be possible to share blood between them and have imperfect seals like with oil and an engine crankcase, but sharing signaling or applying torque would require completely novel methods compared to existing life, as far as my biology knowledge goes.


Wheels and axles did not come from existing biology, unless you are being pedantic. Wheels for locomotion simply was not a viable strategy for evolutionary driven organisms due to their reliance on infrastructure. Why expend valuable energy to build and maintain roads when it helps your competitors? If animals moved with wheels and built roads, the laziest ones benefit the most, and as more and more organisms get lazy, they gotta switch to another method.

I don't think we should worship biology like it's some miracle process with infinite wisdom. It makes sense to look to nature and biology for solutions when the engineering parameters are aligned (see wood, an excellent material that has impressive properties.) Otherwise, a lot of things that we retrospectively attribute to being inspired by nature are merely convergent designs (there aren't many ways to make something aerodynamic, if you would be surprised)

EDIT: I misread your post and I thought you were saying that wheels and axle were from biology. Sorry about that.


The free-rider problem isn't necessarily the dead-knell for biological infrastructure.

Ants and bees build hives after all. Trees build, well, trees.

It's just that for biological systems, it's relatively easier to build legs than wheels. And as you suggest, wheels need more infrastructure to be really useful. So even if you could solve the free-rider problem completely (which you can't), it would still cost biological systems more.


> Ants and bees build hives after all. Trees build, well, trees.

Yeah but those ants or bees are not competitors with each other, neither are cells in a tree. All organism you described have a strong separation between reproductive parts and the ones which build infrastructure.


Yes.

Similarly to how the cells in your bones and muscles don't reproduce.

This is one of the ways biology can work. And it's not in contradiction to what I wrote. Just the opposite.

Though to go a bit deeper: plenty of other creatures derive advantages (and disadvantages) from trees. But for making the biology work, the advantages to the tree genes just have to outweigh the disadvantages to the tree genes.


Wheel spider: https://en.m.wikipedia.org/wiki/Wheel_spider

Earth biology doesn’t seem to have invented powered rotary motion around an axle, that I’m aware of, but I don’t see any reason why it could not have.

And biology has inspired many inventions that have been analogs rather than copies, eg flight.


IIRC, The motor that powers flagellum in bacteria is an example of powered rotary motion.

https://en.wikipedia.org/wiki/Flagellum#Bacterial


Ok that is seriously cool.


There is the issus, a small jumping insect which has developed a gearbox in order to get adequate jumping power: https://www.cam.ac.uk/research/news/functioning-mechanical-g...


Not for power — to synchronize the legs.


Propellers don't rely on infrastructure and they are also absent from biology. Biology simply doesn't seem to be able to evolve such a mechanism.


Below a certain size, propellers are less efficient. It's a fluid dynamics thing. I learned about this from a colleague flying robots inspired by insects. If you find an article about bee flight dynamics that's not BS (creationists still sometimes try to use bumblebees as an example of divine magic), it may also explain it. The short version is that with a small propeller all the air around it just starts spinning.



There are quite a few flagella-like locomotive organelles/glyco-proteins that twist in a corkscrew repeatedly.



While the axel's a bit harder to find examples of, tumbleweed clearly rolls, just like a wheel.



Volta copying the battery from electric fish comes to mind.


This is greatly overstated, though, since the similarity was more structural than anything else. In fact, Volta is just as well known for the much more important demonstration that biological processes were not a pre-condition for electrical generation. From his wikipedia page:

Luigi Galvani, an Italian physicist, discovered something he named, "animal electricity" when two different metals were connected in series with a frog's leg and to one another. Volta realised that the frog's leg served as both a conductor of electricity (what we would now call an electrolyte) and as a detector of electricity. He also understood that the frog's legs were irrelevant to the electric current, which was caused by the two differing metals


> for instance, wheels and axles weren't inspired by existing biology.

Wait, you don’t think “humans ability to roll down a hill faster than they can walk” had anything to do with the wheel?

I don’t know about you, but I had an intuitive understanding of the wheel in kindergarten because that exact phenomenon.


Noticing rolling doesn't require anything biological and the leap from that to wheel and axle is huge and not reflected in biological systems. People were making tools for 10s of thousands of years before they got around to wheels.


Well

1) I suggest you look up the fascinating history of the wheel (as we know it so far)! The axel was actually more likely invented for turning wheels (think pottery); and again, humans are very capable of realizing that angular momentum is conserved when they spin (in that situation they are the axel).

2) I’m a little confused by your point. I concede that no organism has evolved to be Cart-like in its movements, but to say that it had no inspiration from biology seems to create a needlessly narrow definition of invention and inspiration.

3) I think the point you’re ultimately making is that people are clever beings and should use logic instead of simple observation. I think a better example there is nuclear physics (because it very strongly supports your point!)

But maybe, as usual, everyone is right/wrong. Contrary to your point, we are constantly finding new biological inspiration for invention (from nano to macro scale!), but to your point, some of the coolest stuff we do could never come out of the biological Monte Carlo simulation we are a part of.


1) No, a rolling thing is not a thing with an axle, no amount of looking things up will change that.

My overall point is that the bulk of human technology is not very strongly inspired by biological systems. We'd certainly like to build machines with some of the neat properties of organisms but until very recently we simply didn't have anything resembling the capability. Most machines are designed in principally different ways - modularity on a different scale, different use of materials to achieve reliability, etc. Wheel and axle is a just one of the most basic examples.


Golly, maybe to clarify what I’m saying.

Wheel invented for pottery spinning. Pottery spinning machine very similar to human spinning. Probably inspired by human spinning

Human now has spinning machine. Human sees that when they spin down hill, they move easier. Human tips over pottery wheel.

Why is the above such a hard thing to believe? It’s supported by the archaeological record! You may not like that narrative but what’s the one you have?

And to be critical of the point youre making: there’s only so much chemistry and physics out there and ultimately we are made of biology, so what are you even saying? That we love making symmetrical objects but that has nothing to do with biological symmetry? That we forge metal because we are attracted to shiny things but that has nothing. To do with our obsession with looking at eyeballs? Maybe open your mind a bit?


I always imagined that early humans had fun rolling big rocks or logs down hills (who wouldn't?) and someone got clever.


Maybe!

But our current findings imply that the wheel and axel were invented for pottery turning and someone eventually got the cart thing going sometime after that.


> wheels and axles weren't inspired by existing biology.

That suggests the hypothesis that only universally discovered mechanisms (which does not include the wheel and axles) were inspired by natural phenomena.


What's a universally discovered mechanism?


The fist electric batteries via electric eel's electric organ's cell stack.


The history of aviation is full of people taking inspiration from flying animals


It's a very broad inspiration and the ones who most closely hewed to biological systems were some of the least successful - most flying machines don't use active wings with highly variable geometry. The first successful human flying machines didn't have a meaningful biological equivalent at all.


Depending on your perspective, the airplane is not successful until we figure out how to do it sustainably. We may yet have to go back to lighter than air flight, which arguably has precedent if you count aquatic life, i.e. swim bladders.

Unless we start making the sustainability of our technology a success criteria, we'll never achieve it. Many biological systems are different precisely because they are constrained to run within an overall sustainable system.

Maybe human tech ~is cyclically sustainable on a long enough timescale, but I certainly ~hope we're not going mostly to die off and live without technology until the forests grow back and microorganisms in the oceans turn into oil.


Sustainability is not something optimized by natural selection either, all that matters to get a trait adopted by a population is that it provides a reproductive advantage to the individuals with the trait, which has frequently in the past resulted in ecological collapse. The most disastrous and classic example being the first organisms to emit oxygen which at the time was a deadly poison, is assumed to have killed a huge percentage of the species on earth, and could easily have resulted in a genuine full wipe.

We're smart enough that we should be able to avoid anything like that if we put in the work but don't glorify what's natural, we should be looking for our own solutions.


The three-axis control system (in particular the concept of ailerons) developed by the Wright Brothers was inspired by the way birds twist their wings.


And then plummeting to their death, right?


It's a process. Darwin talked about it.


The process being that the people who took inspiration from animals died out and gave way to the people who were better able to adapt to changes in modern technology like the propeller and internal combustion engines?


Yes, thank you for clarifying the joke :)


The ideal tool to "discreetly capture [insects] for sample collection". At least 700 actuations before it starts degrading, that's impressive.


Is using the corpse of a bug predator really a good way to capture bugs? It seems like that’s something they’d be likely run away from.


Most insects can't really see that well and spiders don't hunt the way the scientists are using them, like a claw machine. I would guess it won't set off their normal defense alarms.


Biomimetic ideas in engineering go way back. In Greek mythology, the Daedalus story starts with him getting trouble for killing his nephew Perdix out of jealousy - the younger man had invented the saw by mimicking the ridged pattern of a fish skeleton (or serpent's jaw) in the edge of an iron sheet.


If I got it right, everything about the spiders was already known, and the only thing suggested in this research is to stick a needle into one, so you can extend and retract its limbs. It doesn't seem too innovative, and very likely they weren't the first to think about it. If they found a way to extend the duration for which these spiders can be used, that would have been more impressive, but even though they made a suggestion, they didn't pursue it.


They weren't the first to think of it. I wondered this as a child in the 90s, as did my father once I mentioned it. Our conclusion at the time: useless. The cadaver will degrade extremely quickly, thus its usefulness in a factory versus a small metal-based mechanism that has a MTBF measured in months to years is always preferred.

Modern pick-and-place machines are far better than any hugely variable cadaver, and decent quality ones will work for a very long time with little maintenance.


Without even realising it was a brag until now I had this idea the moment I heard spiders were hydraulic and dismissed it as silly haha

Should have been a scientist


So... something reminiscent of Pickle Rick, from Rick and Morty? (https://en.wikipedia.org/wiki/Pickle_Rick)

Spoiler: Rick lands himself in a pickle, and "lacking any means of mobility, he bites the head of a cockroach and walks upon its back by stimulating its brain with his tongue. After assembling more cockroaches into a crude exoskeleton, he sets up a lab and upgrades to a powered exoskeleton made of rat corpses."

Fun episode!



I saw some videos of this on TikTok, and I still can’t decide if it is awesome or horrifying.


It's very "Are we the baddies" research.


Another example, I've seen work where it was proposed to essentially stick a small Geiger counter in the pleasure center of the brain of a flying cockroach.

Idea being that you release them and they'd swarm to someone smuggling nuclear material.

I think some actual demonstrations were done iirc. Can't find the abstract for it right now


They even call it "necrobotics". They're creating undead robots. They're pretty much recreating one of the worst sci-fantasy horror scenarios.

Still fascinating, though.


Live in the pod, eat the bugs, and use the bugs.


It is fascinating, yet it still makes relatively little use of a complicated and capable machine that is a spider, relative to when we can be in a few decades from now.

On the other hand we still can’t even get dogs to deliver mail, so…


I don't understand what spider legs are doing that we can't manufacture ourselves. I don't really get what problem this is solving. Just seems kind of fucked up to me.


It works because of the "inherent compliance of the legs as well as hairlike microstructures on the legs that work kind of like a directional adhesive". It would be hard to manufacture a tiny pressurized system with valves and those hairlike microstructures. Spiders are self-assembling and made out of cheap materials. Anything with piloted corpses does seem a little fucked up. I'm creeped out imagining accidents storing live spiders or weird research advancements on engineered spiders.


I expect the same line of thinking by the aliens repurposing our brains :)


From the first paragraph

> making a robot that’s an insect equivalent is extremely hard—so hard that it’s frequently easier to just hijack living insects themselves and put them to work for us.


Doesn't seem easy to scale. If you wanted to build a factory or something with this you are going to need enough to make it worth investing in building them in a reliable and consistent way.


Building a farm to deliver 10000 spiders is orders of magnitude cheaper than building an assembly line that builds them. All the spiders need is a food source and an appropriate temperature/humidity.


That gives you a box of dead spiders, but surely the hard part is everything after that like evaluating corpse viability and fitting tiny spider corpses to input sources.


> fitting tiny spider corpses to input sources

Don’t you have to fit tiny artificial spider legs to input sources too? That’s not a unique step


That's my thought exactly. Handing dead spiders and installing hardware isn't exactly easy.


We wrap ourselves in the skins of slaughtered animals, line our cars with them, cook and eat their flesh, suck their milk out and bottle it to drink, perform experiments on live mice and monkeys injecting them with chemicals and experimental medications at increasing dosages to find side effects and death, spray toxic chemicals that cause a holocaust of invertebrate life over millions of acres of farmland, harvest millions of tons of fish and "bycatch" out of the oceans by dragging nets through them.

Playing with a few dead spiders is pretty mild in the scheme of things I would say.


Biology and culture has prepared us for eating other biological beings, so steak and fish aren’t gross to most people.

For most people, to use a non-food animal product, it has to be significantly unlike the animal itself, eg it’s easy to enjoy leather or fur that is not in the form of an animal.

Even for food, most people are less comfortable with food the more like themselves it seems, eg a pig head on your plate is much less appealing than a pork chop. Lobsters and shrimp are not a lot like people so the disgust factor is significantly less.

Yes, there are people with morbid interests in (actual, not depicted) skeletons and taxidermy but these seem to be getting more rare in modern cultures.

For whatever reason, spiders trigger disgust and/or fear in most people; I can’t see necrocyberspiders ever being popular.


Yes biology and cultures has also prepared us for squishing spiders. I assume the OP is calling this fucked up on a moral or intellectual level, rather than the normal basal dislike of spiders.


> for whatever reason

I’m inclined to think this is instinctive, as some insects and spiders can be dangerous to us, we are instinctively repulsed by them


Aww sweet, manmade horrors beyond my comprehension


Eerily similar to "Roach Biobot".. I'm very surprised this was published, the cockroach is still alive.

https://phys.org/news/2012-09-technique-remotely-cockroaches...


I have seen this as a major news headline for almost a week now.

is it just me or is this just stupidly juvenile? We've known that spiders are hydraulic forever (its why they curl up when they die lack of pressure).. how does toying around with their copses make mainstream headline news?


There was this one case where someone made a quadrocopter out of their dead cat[1]. That was also in the news, while we gained no new knowledge from it. "Look at them, they play with a corpse" is quite newsworthy, apparently. :D

At least in this case, we now know for sure that a spider-powered UFO catcher / claw crane is possible. :P

[1] https://www.dw.com/en/the-dead-cat-drone-with-an-eco-message...


It's not just you. I have no idea how this is interesting or useful. Plus it's pretty fucked up.

This, along with the meat article, makes me wonder if the IEEE has lost their mind.


So this is how the skeksis created the garthim


Did the spider donated his body to science?

In case he’s didn’t, I hope the scientists asked a permission from his first-degree relatives.


what is that supposed to be useful for exactly? I can't really wrap my mind around it tbh.


fuckin horrific


My new startup uses crowdfunding to pay Jeff Goldblum to visit people and deliver the line about being so excited by the question of whether they could, they forgot to ask about whether they should.


Like reading a garden-path sentence, I really thought you were going with The Fly and I had to backtrack to Jurassic Park.


I assume dogfooding the concept came naturally?


Yeah I let him talk me out of it.


"pneumatic": so basically they apply pressure by injecting air into spider's valve and it moves?

How is "necrobotics" a real field of research? :p


step 1 - convince someone at DARPA you aren't making an army of robot zombies


Or suggest to them that you are indeed making an army of zombies, but that they’re super patriotic zombies.


And that the "zombie gap" is a growing concern...


No no no, convince someone at DARPA that you can be relied upon to deny it in public.


That it's vital that the US develops these robot zombies before Russia/China does.



I don't like it


The paper says hydraulic. pneumatic != hydraulic


Their embedded video mentions they use a puff of air, but I don't see that in the paper so maybe the video title just got it wrong?

https://www.youtube.com/watch?v=1JOS6hMHIUM

Or I guess it still may be air in the syringe pressing on fluids in the spider and be kind of both?


I find killing of animals really offensive, haven't we killed enough on the earth?




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: