Slashdot Mirror
Like A Cat, New Robot Lands On Its Feet
eckenheimer writes "Students at the Physics Department at Drury University have developed a robot that uses motions and contortions of its body
to orient itself in zero gravity. According to the project site, 'If you've ever seen a cat land on its feet after falling while upside down then you've seen the idea behind our project.' The effort is a proposal for the NASA Reduced Gravity Student Flight Opportunities Program."
That sounds like a fun project to work on. I can think of all sorts of uses for something like this. We can ensure that all olympic divers enter the water perfectly perpendicular to the surface. likewise gymnasts doing the vault will always land on their feet. Throwing spirals with a football could be automagic. Ok...nothing lifechanging there...but I'm sure someone will think of something.
Better than Flickr - Manage, Share, Archive
When a cat is falling ( or while anything is falling for that matter ) it is in free fall until there is significant drag from its motion through the air. Free fall is effectively a zero gravity state.
NASA used planes in a dive to simulate zero gravity for astronaught training.
Actually, it's the short falls that tend to kill cats. Cats (like skydivers) can assume a position that reduces the terminal velocity and presents the greatest surface area for impact, reducing the force per unit of surface area. It takes a while to rotate and get into the position, so if the fall is too short, the cat will land in an awkward position and is far more likely to die.
This is not to say that the cats that fell from a great height were uninjured - just that they were more likely to have non-fatal injuries.
This sig seemed like a good idea at the time....
...if the robot can survive terminal velocity falls like cats. Cats falling from very high heights (i.e. skyscrapers) tend to survive the fall better than those falling from lower elevations.
There are serious applications for this. Namely, being able to orient yourself in space without using propellant is a very valuable ability.
The real surprise is that this is nothing radical; I really don't get why it deserves a press release. Probes have been using momentum wheels, for example, to do the same thing for decades. What is the big advantage of this over, say, a small set of momentum wheels? All I can really say about it is that it's more complex, and seems more likely to wear and have part failures.
Very well; let this abomination unto the Lord begin!
In grad school one of my physics professors wrote a paper on orienting onself in zero g with no net angular momentum. One student was just convinced it was impossible. Soon thereafter we were visted by Story Musgrave (one of astronauts who fixed Hubble) and the professor told him of the paper. Story immediately sat down on a swivel chair and demonstrated the motion necessary to turn in zero g without grabbing on to anything. It's interesting how a concept that caused some interesting debates among the students suddenly became obvious when it was directly demonstrated.
...they should add the technology to this robot?
Since it's a one-axis device, there's no need to test it in a zero-G environment. Hanging it from a string would work equally well.
There's useful work to be done on three-axis stablization algorithms, but this isn't it.
Even NASA didn't go for this one.
No actually, the paper he cited is not the only research that has been done on this - I distincly remember reading an article about this 3 or 4 years ago, and a radio program mentioning it a while back too. I didn't read the paper he pointed to, but the reason it works is because not only do the cats orient themselves properly, they also splay out their legs and stretch the skin out, creating a parachute-like effect which drastically reduces their terminal velocity. Cats falling from 5 stories have plenty of time to orient themselves, but not enough to get this parachute thing going and slow themselves down (i.e. they're falling at the terminal velocity of a bunched up cat, not a spread out one). Also, you seem incredulous that cats falling from over 8 stories have a 100% survival rate. Remarkably, in reality it IS almost this high, certainly >90%.
Cats survive a one story fall better than two or three; they only require about a metre's fall (or less) to right themselves and extend their legs. I've seen my little cat do it in about a foot and a half when she rolls off her scratching post platform, but she's small and fit (it's actually particularly impressive when you consider that she's usually stoned at the time ;).
The paper considers all heights over seven stories to be the same because a typical cat reaches terminal velocity in about five stories, at which point it detects the end of acceleration and goes spread-catted, requiring another couple of stories to decelerate as much as possible. Thus the really dangerous height is five or six stories.
Cats certainly can and have survived great falls - they'd never have evolved this behavior otherwise. They have many other advantages and adaptations as well, like their lack of shoulder joints, generally impressive and flexible musculature, and their ability to ragdoll for an impact (similar to the way a drunk person has a better survival chance once a car has crashed).
There is no implication of a 100% survival rate at great heights - you're simply misreading the data, and the conclusion. You may want to consider this likelihood in future when you see ludicrous implications of a study.
I've dissected a cat. It was pretty much a standard short haired cat. I think it must have been a stray alley cat, but not one of the bright ones that was smart enough to run like hell when the cat-snatchers came.
Anyway, Once you see a cat without it's skin, the reason that cats can take falls becomes apparent. the only really massive structure in a cat is the legs/shoulderblade/pectorals structure. The shoulder blades on our cat were huge and had an endless number of muscle attachments. The shoulder blades are hooked on to the back but the connections are relatively loose and sloppy and the spine is basically like a slinky. The legs on our cat had a 5+ inches of travel. If you get a chance, pick the cat up by holding it with one hand under it's ribcage. Use the other hand to work the front legs up and down to see the massive travel and check out the way all the muscles that insert into the shoulder blade take up the shock. Giving the cat a tracheotomy and watching the lungs inflate and deflate is a lot of fun too.
The hind quarters of the cat weren't really much, but it was as it was a really skinny alley cat there wasn't much weight for the back legs to handle. The above poster's postulated "enormously fat, bald, tailless" cat would probably shoot it's guts out all over the place on impact, but a normal cat might have a chance. Also a long tail might be handy for balance and steering on the way down, but I don't know.
Take A&P if you ever get a chance. It's a lot of fun. But try not to get hooked up with idiot lab partners that only enjoyed dissecting the cat's nuts. That freaked me out a bit. Also, don't wear your good shirt to a lab session.
Why do I have this? I don't smoke.