Swimming Cockroach Robot Developed

Onnimikki writes "The Ambulatory Robotics Lab at McGill University has made a six-legged swimming cockroach robot as part of Project Aqua. The robot is a waterproof version of the RHex robot, whose inspiration is the biomimetic work by Bob Full of Gecko glue fame. Other cool stuff from the ARL page includes a waddling bipedal RHex, and the world's first galloping robot."

Comments

Fun

[Cackmobile]

I could have so much fun with one of these. Scaring my mum/sister/girlfriend. I want one!

Re:Fun

[billybob2001]

Scaring my mum/sister/girlfriend.

The scary thing is, you're talking about one woman

Re:Fun

[Cackmobile]

nice one!!! it should have been mum, sister or girlfriend

Re:Fun

[lmfr]

hm? "/" means "or", doesn't it? ("or" as used in common language)

Re:Fun

[56ker]

Yes, as in all those letters you'd get home from school (when I was at a co-educational school) which mentioned the phrase son/daughter about five times in the letter. That was the 80s though - I'm sure these days you could link it to the gender field of a student database and print out the correct one. It'd be a bit more complicated than a mail merge though.

Re:Fun

[Anonymous Coward]

Remember Abort/Retry/Ignore?

Did you press A,R and I, or just choose one?

Re:Fun

[basser]

and I have to suffice with my g/f stepping on real cockroaches.

Re:Fun

[cap'n foolsy]

this from a guy whose name is billy bob?

Re:Fun

[Anonymous Coward]

I, for one, welcome our new aquatic cockroach overlords.

Re:Fun

[Anonymous Coward]

Cute but not quite possible. Mom couldn't give birth to herself as a sister. I'd suggest that mom is his girlfriend and that is as good as it gets.

~S

Re:Fun

[BrokenHalo]

Dunno about your roachos, but the ones here in Perth (Western Australia, that is, not Scotland) don't swim too well at all. Just about the only way to kill the bastards is to drown them...

Re:Fun

[Anonymous Coward]

His mom/girlfriend is his sister in the metaphorical sense not the literal sense.

Re:Fun

[Anonymous Coward]

But sister could give birth to poster fathered by her own father. That would make the first two (mom/sister) possible. The girlfriend part would have to happen later. Ewww... I'll shut up now.

Sticky situation

[billybob2001]

inspiration is the biomimetic work by Bob Full of Gecko glue fame.

Why is he full of Gecko glue?

Talk about getting stuck into your work...

Re:Sticky situation

[SEWilco]

inspiration is the biomimetic work by Bob Full of Gecko glue fame.

Why is he full of Gecko glue?

You don't know the story behind that?
I thought everyone knew.
The way he tells the story is what really made him famous, not his being full of it.

It runs QNX

[leeroybrown]

It's nice to see that it runs a proper Real Time OS.

I have actually seen one case of someone trying to build a mini sub-aqua robot running Windows XP (yes XP not CE) on a powerful micro PC card.

Seriously, ... it sounds fscked up, but it's true.

Re:It runs QNX

[Rosco P. Coltrane]

Using Windows CE or XP as a real-time OS is possible, it's just a matter of warping time. Time warping isn't possible yet, but the Microsoft RTOSes that use this feature already exist. That's how far ahead of the curve Microsoft is.

Re:It runs QNX

[d3faultus3r]

I think I know what happened to that robot!

Easier way

[Anonymous Coward]

They could have just asked me, I've got plent of cockroaches around here and could have lent them some.

Buoyancy

[Rosco P. Coltrane]

waterproof version of RHex, which was made neutrally buoyant

How did they do that ? Depending on the density of the water you immerse the thing it, they might approach neutral buoyancy by adjusting the amount of ballast manually, but they'll never achieve true static buoyancy without some kind of active process controlling the amount of water in a ballast tank. Otherwise the object would sink to the bottom or bob up to the surface eventually. Or do they maintain the thing's depth in the water with dynamic buyoancy using the robot's forward movement ? I don't see depth control planes on the robot, could they use its legs to achieve this ?

Re:Buoyancy

[Coelacanth]

You can make an object neutrally-buoyant (or close enough to it) by carefully adding foam or other light stuff (ping-pong balls!). The tricky bit is making it not only neutral in an overall sense, but to prevent the object from tending towards a particular attitude in the water.

And unless you fill the tank with salt water or, perhaps, lime jello, the density of water is pretty much the same everywhere :-)

Re:Buoyancy

[Anonymous Coward]

Erm... huh? Since when ? In water, which is nearly incompressible, the density is the same regardless of the pressure (well it's not, but it's totally negligible compared to the changes in density due to the salinity of the water).

Go back to school.

Re:Buoyancy

[Surak]

And unless you fill the tank with salt water or, perhaps, lime jello, the density of water is pretty much the same everywhere :-)

Why particularly *lime* jello? I'm allergic to limes!

Re:Buoyancy

[aaarrrgggh]

Water is incompressible; the only potential impact on neutral bouyancy is if the shell of the robot can deform, thereby reducing its total displacement. (Foam or ping-pong balls would have this problem, if used for neutral bouyancy.)

Re:Buoyancy

[Zog The Undeniable]

OK for shallow depths, but as the robot gets compressed at greater depths, its volume decreases and it will sink faster and faster (like a submarine that blows too much air from its tanks). There are also problems with temperature change. There's an interesting passage in Lothar-Gunther Bucheim's book "Das Boot" (better known as a TV series and film) where the Chief Engineer explains the variable buoyancy problem caused by temperature, depth and varying salt concentrations to the narrator. The U-boat has to take on or lose a surprisingly large weight of water to compensate for even a 1 degree change in water temperature.

But I digress, I doubt there's much call for a deep-sea robot cockroach.

Re:Buoyancy

[FreakyDeaky]

Could the Robot's body itself be the depth control planes, I mean like by tilting the things body using moving weights or something. I'm by no means an engineer so i was just wondering if this was possible.

Re:Buoyancy

[Pvt_Waldo]

Don't forget it is the density of the vehicle that counts, not it's weight. So instead of changing it's weight, you change it's volume. Easy to do if you have some kind of piston which has one end open to the water. Move the piston out, you increase the volume of the vehicle. Move it in, decrease the volume.

Re:Buoyancy

[anactofgod]

You don't see "depth control planes"?

Really? I see six of them. *g*

Well, I guess they technically aren't "depth control *planes*", but by altering the angle thru which the various flippers oscillate in relation the the body, the roach can be made to level, dive, rise, roll, etc.

Watch the whole clip and prepare to be amazed.

Having participated in a previous life in the building of underwater robots to test system, designs, and procedures related to space station maintenance, I can tell you that making a robot (even a large robot) neutrally buoyant (and "attitudinally/orientationally neutral") may be a real pain in the ass, but it's hardly rocket science. Especially when that robot is being operated in a controlled environment like a neutral bouyancy facility (ie "swimming pool").

...anactofgod...

I'll make a mint

[snatchitup]

With a little hack. I'll be able to retrieve all of Colin Montgomerie's (Monty)'s balls from the lake at the US Open starting today!

Come on Justin!

Bill G's cunning pest control plan

[Snart Barfunz]

Robot cockroaches running Windows infect the bugs with bugs. Insect population crashes, MS share price soars.

Not first

[quintesse]

Wish I could prove with u URL of some sort but I'm 2000% positive that I've seen galloping robots in a documentary years and years ago made by some university or MIT. I remember that they were made using hydraulics and that they had quadrupeds and even a monoped running/hopping through the hallways (with the researchers running to keep up with cables and such ;-) I also remember that the movements were not preprogrammed but the system "learned" how best to cope with N legs. It developed all of the gaits found in a horse for example. Very good stuff.

Re:Not first

[Anonymous Coward]

Yes, I remember seeing a documentry on Discovery Channel about the MIT team doing what you said.

Re:Not first

[pipingguy]

This thing almost loooks like it could gallop (walking tree harvester from Finland - apparently it can sidestep like a crab, too).

Website

Re:Not first

[Onnimikki]

You may be 2000% positive, but the assertion that no galloping robots had ever been made (until now, by MIT or anyone else) is backed up by Schmiedeler and Waldron's IJRR paper entitled "The Mechanics of Quadrupedal Galloping and the Future of Legged Vehicles". In it they state "To the best of the authors' knowledge, however, no artificial legged system has ever been operated in a true gallop. Raibert's (1986) quadruped used its legs in pairs, employing trot, pace and bound gaits." The MIT work that you are referring to is that done by Marc Raibert.

Re:Not first

[Anonymous Coward]

The robots you are talking about are on this page. Check out the running and hopping robots halfway down the page.

Re:Not first

[aka-ed]

How do we know that even this robot is capable of a "true" gallop? Have we asked a horse about this? How about a "Gallop" Poll?

Eeaaaargh!!!!

[BabyDave]

*hits cockroach with shoe*

What? Why is everyone looking at me like that?

Re:Eeaaaargh!!!!

[Anonymous Coward]

*hits cockroach with swim fin*

Is that better?

build your own

[Anonymous Coward]

http://www.lynxmotion.com/

original rhex is equally impressive

[Anonymous Coward]

Yes, the swimming robot version is cool, but the original rhex robot is pretty incredible too. I work in the lab at UMich where they're working on the land-based one. A friend of mine used a learning algortihm called Amoeba (sort of a hill-climbing approach using simplexes) to speed it up dramatically. It runs fast, much faster then you would expect a stocky little robot with six legs to run. Currently, they're working on a vision system so it can track objects and follow lines, and having it sense its terrain and modify its gait accordingly. Not your daddy's robot!

Re:original rhex is equally impressive

[SEWilco]

Currently, they're working on a vision system so it can track objects

It doesn't really need to track objects to be able to run when the lights go on until it is in a dark place.

Waiter!

[da3dAlus]

"What's this robotic cockroach doing in my soup?"
"The...um...backstroke?"

What The World Needs Now

[SEWilco]

I want a robot which can fetch me a cold drink, not swim in it.

Already done better

[Anonymous Coward]

In this year's Technogames (in the UK, broadcast on TV by the BCC), there was a robot that could swim underwater. It swam like a fish, with horizontal tail movements, and knew when to move up or down to stay slightly below the surface of the water. It was autonomous, not remote controlled. Much more impressive than the movie of the swimming cucaracha.

Re:Already done better

[FreakyDeaky]

I saw that also. Didn't they have plans to develop submarines and boats with that kind of propulsion instead of screws?

Re:Already done better

[uberdave]

That one would have problems when it got onto land. This bot could be made amphibious by stiffening the "flippers" enough to support its weight on land.

Complicated much?

[pitboss8881]

Whatever happened to the wheel? You know that wonderful invention that converts rotational motion into linear motion. Hey, our offroad vehicles use it. Are our robots too good for such antiquated ingenuity? Is the answer simply too easy to give the robotics community the type of intellectual hooplah they thrive on? Or is there some technical reason why trying to make a robot walk is better than letting it roll?

Re:Complicated much?

[Anonymous Coward]

>is there some technical reason why trying to make a robot walk is better than letting it roll?

Walking machines will always be able to reach more places then rolling robots.

However, the faster robots will (should) always be the rolling ones.

Re:Complicated much?

[Anonymous Coward]

It is too easy to program wheel-based robots. They seem to be used more frequently for following black lines on the ground, object avoidance, and following walls - things like that.

On the other hand, walking robots often have multiple servos at each leg. Thus, in order to take a few steps, it requires a complex progression of commands that cause each leg to lift, rotate forward, and so on.

loop thru legs 1 - 4
Shoulder servo: up 45 degrees
Elbow servo: up 30 degrees
end loop
loop thru legs 1 - 4
Have servos return to neutral position
end loop

All that just to take one step.

Once you have the basic walking mechanics down, you can invent different paces, gaits, and variations on the theme. You can use a controller chip such as the basic stamp to program all this stuff in (http://www.parallax.com).

Then, once you have defined 'walking' from a programmatic standpoint, you can use an RC adapter to send commands to the basic stamp, which interprets the signals as forward (run the embedded walk program we defined above), backward, etc. Much more interesting than rolling about!

Plus, alot of these robot guys have BattleMech fetishes. They won't quit until there are Ullers running around the lab.

Re:Complicated much?

[Ami Ganguli]

From what I could figure out before it was Slashdotted, the advantage of legs is mobility. RHex , for example, can run over obsticles that are the same height as it.

Wheels are limited to obsticles smaller then the radius at low speeds, and considerably smaller than than that at high speeds. No idea what the actual formula is though - anybody know?

Re:Complicated much?

[LeoDV]

Probably the same reason we evolved with legs and not wheels. :p Legs are much more handy. A robot on wheels wouldn't be able to go up or down stairs (which will be required when they're used domestically), will be less manoeuverable/adaptable the topology of his environment, etc.

But yeah, you have a valid point. Actually teaching a robot the complex algorithms it takes to walk on legs always was a challenge waiting to be tackled, and I'm pretty sure the challenge element was part of the decision process.

But if I had to pick a deciding factor, you have to keep in mind that the entire field of robotics is aimed towards, in the very long term, making androids, that is human-like robots (I think we've all seen that robot from MIT (?) who is capable of expressing facial expressions through the use of flexible plastic eyebrows and lips), and not the ones we already have to build cars, and that includes walking around on two legs.

Re:Complicated much?

[FreakyDeaky]

I really have nothing to add except that would freakin rule!!! (as you could tell I have a passion for battlemechs)

Re:Complicated much?

[OwnedByTwoCats]

A robot on wheels wouldn't be able to go up or down stairs

Dean Kamen doesn't think so.

Re:Complicated much?

[nvalid]

There are at least two advantages to legged robots that I'm aware of (though the technology is not necessarily there to take advantage of them). The first is the potential to climb much larger obstacles for a robot of a given size. Robots using rocker-bogie wheel systems such as the Mars Pathfinder vehicle have amazing climbing abilities, but this is nothing compared to what a human can accomplish.

The second is a potential energy savings. Imagine a wheeled vehicle traveling over rough terrain. It's constantly climbing over obstacles which takes energy that is just lost when it falls down the other side. Meanwhile, a legged robot can keep its body above the height of most obstacles and just step over the top of them -- more of its energy goes towards its forward motion instead of the up-and-down motion of the wheeled vehicle.

Oh, another thing is the ability to tolerate loss of an actuator. If one of the wheels were to stop working on a wheeled vehicle, the rest of the wheels would have to drag that one along. Meanwhile, there's been some neat work showing the robustness of legged robots to such problems by groups such as the Biorobotics Lab at Case Western.

In the end though, it depends on your application as to which is best. I just can't see that one approach could be better than the other in all cases. Just as one example, I think legged robots have really cool potential for planetary exploration for the reasons given above, but certainly anything spending most of its time on flat ground (agricultural equipment, anything on-road, etc) would perform better with wheels.

Re:Complicated much?

[Anonymous Coward]

Ullers? Come on. That one's only 30 tons! We all know that what they really want is a Dashi!

Re:Complicated much?

[Dylan Zimmerman]

Robots with wheels certainly can go up and down stairs. They just have to use a strange arrangement of wheels. Specificaly, triangles of them with one wheel at each vertex. When one wheel encounters a pothole (or in this case the gab between two steps' edges), the entire assembly rotates. It works rather well. I've seen a robot designed like that climb about 20 stairs in the same time that it would take to go about 15 feet. The rotation slowed it down a bit, but really, it was pretty efficent.

And the robot from MIT that can do facial expressions is called Kismet. Quite a cool idea, but it can't do much without a mobile base. Right now, it just sits in a room and waits for people to interact with it.

In my opinion, Cog is much more impressive because it learns. They showed it how a toy car works and it learned how to move the car very quickly. Cog actually learned that the car would only move if it is pushed on the front or back, not the sides. Now that's an acomplisment.

Re:Complicated much?

[Dr. Smeegee]

On a large machine that must pick it's way carefully through a delicate area, legs may be a decided advantage.

Scope ye the Plustech Harvester. It uses six hydraulicly actuated legs and a big-assed arm with a gripper/saw on the end to harvest trees from delicate areas.

Some guys I went to high school with use horses to harvest hardwoods from Indiana forests due to the trees inaccessability to wheeled vehicles.

Legs are useful!

Heh...

[emo boy]

you said mum...

Excerpt from Meet the Parents:
"Don't you have any expensive wine?"
"hmmmm....you could buy a bunch of Mum's"

I've been slashdotted! YEAH! I'm so proud!

[dannycim]

Came in this morning, tried to login to squirrelmail.... Hmmmm... very slow... Get to the web server.... Hummmm... Lotsa httpd processes... Hummmmm...

tcpdump -i eth0 -n port 80...

Hmmmm... The console scrolls non-stop! Arrrgh! Am I being DOS'ed!?!?!?!?

Thanks slashdot, you made me panic for a while. Hope somebody mirrored the pages cuz' I can't handle this load without being prepared for it.

Please check again in a few days if you're really intereseted.

--
Danny, McGill CIM SysAdmin.

Re:I've been slashdotted! YEAH! I'm so proud!

[cameronsto]

How would you have prepared for this?

WinXP and the newbie Roboticist trap

[snatchitup]

Many people start robotic projects fearing embedded development. So, they think, why can't I just control everything from my PC.

The problem with this is, it actually adds complexity.

Typically, it means adding a MAX232 with Charge ups, or the more expensive MAX233. This, just to convert the RS232 25Volts down to TTL 5volts. Then you need another component to translate the characters into logic. What a pain! Not to mention a tether.

Better to just learn a little assembly. It's really easy for these applications. Just turning things on and off is setting/clearing a bit in an output register.

Software, is really not that hard, in fact, possibly overrated in terms of the complexity of building one of these beasts. It's the electronics, and contruction. Getting things to actually move.

Re:WinXP and the newbie Roboticist trap

[apdt]

True, especially when you can actually program some of the PIC's in C. e.g. the Microchip PIC 12/14/16/17 families

Re:WinXP and the newbie Roboticist trap

[LinuxInDallas]

If someone is developing a robot the absolute least of their complexity worries is an RS232 transceiver. There's nothing wrong with coupling your home built robot to the PC. That at least lets you send it data and likewise allows it to send you data. Wireless is also an option with the low cost modules that are now available.

As far as having a processor on the robot, obviously if they are sending RS232 data down to the robot then they must have a processor resident on it that is interpreting those commands.

Re:WinXP and the newbie Roboticist trap

[juhaz]

Not only PICs. Lots of other microprocessors generally used in embedded projects are C-programmable as well.

A galloping robot?

[Prince_Ali]

Did that trigger thoughts of the Animatrix for anyone else?

JUST ONE THING TO SAY

[Docdawolf]

You have me hook i just have one thing to say where can i buy one at? It kinda reminds me of Blade runner I would love to own one the fun I would have at the beach or pool in florida.

Re:I've been slashdotted! YEAH! I'm so proud!

[yuri82]

he would've kept a fire hydrant by his server

Re:I've been slashdotted! YEAH! I'm so proud!

[dannycim]

I'd have prepared for this by mirroring the images and videos and redirected to them. Ain't so hard if you know in advance.

I didn't post the story, somebody else here at McGill did without telling me.

Anyways, anybody want to host 'em?

Great, More Pests

[fuzzybunny]

Just great. Now I have to keep a portable EMP generator next to my cans of RAID under the kitchen sink. Do you know how much power those f***n things use?

Let's hope they don't teach the little bastards to breed. What will they do, lay their eggs in my box of spare PC parts? Although I assume there'll be a nice satisfying mechanical *crunch* when you step on them.

Gives new meaning to cockroaches carrying diseases. Maybe they'll find one that transmits W95/Klez@mm. Norton Antivirus will now cost three times as much to ship, because it comes with a large hammer. Don't download files, don't open mail attachments, and put a ring of flea powder around your PC. "Dr. Solomon..." *WHAM WHAM WHAM* *crunch* "...has detected and isolated a virus."

On the other hand, it'd make for a nice way to smuggle an X10 cam into a cute girl's bedroom--assuming they ever make the transmitter units weigh less than 5 pounds. Blattidae Elegans Electronicus indeed.

Movie mirror

[Onnimikki]

I've posted the smaller movie [14MB] on the .Mac servers: the cockroach robot movie.

Quick, Henry- the Flit!!!

[SolemnDragon]

I do not want a motor roach;
I do not want them near my boat...
It's bad enough on Land and Beach-
Keep that insect out of reach!!!

I do not want electric roach,
Nor the real thing to approach!
Don Marquis, entombed, is spinning-
And now, you say, the roachbot's... swimming?

My house is currently invaded-
I didn't want this thing created!
Why not start with robot ants,
or with spiders, tech advance-

I do Not LIKE the cockroach breed!
I do not want them close to me!
I'm scared of those alive and well-
and these
not even Flit can Kill!

My morning tribute to Dr. Suess and his earliy career, and Mr. Marquis, whose cockroach vers libre poet would have had quite a bit to say on this one. Those were the first things to leap to the forefront of my mind- Quick, henry- the flit! and oh lordy what would archy say.

I'll have to develop...

[dentar]

The robotic pointy boot to step on it.

At the End of the World

[WC as Kato]

Great, after the nuclear holocaust, there will be nothing on Earth but a few humans fighting off highly evolved Terminator-like robotic roaches and radioactive cockroaches.

Re:Ok so it's like this then?

[Anonymous Coward]

Yes sir (or mam, as the case may be)! Good work!

Ahh now i can sleep at long last

[Denver_80203]

Thank god someone as finaly developed a swimming cockroach robot. Our only lingering question is: what more is there to accomplish for man kind? Now tha the swimming cockroach robot hurdle has been cleared... what goal is wirthy of a race? IS it a happy day... or sad?

here it is: bittorrent file

[nevroe]

I work on the RHex project at CMU, so I had the movie sitting around on my computer. The ARL website is pretty dogged down, so James had to pull the website and the original 60 MB video, replacing it with a smaller 14 MB version.

If you want the full version movie, go here for the torrent file.

http://www.cs.cmu.edu/~gch/Aqua.mpg.torrent

This is Slahdot...

[Spiked_Three]

Why hasn't someone already accused Microsoft of inventing the swimming cockroach long ago?

Backwards

[nuintari]

Damnit, we are trying to kill roaches, not make more of them!

And waterproof, next it will be squishy proof. I find these anywhere near my house, your gonna pay to have em exterminated.

Robot video page. Wooo!

[azav]

You guys GOTTA go here:

A "pronking robot"
http://kesisleme.eecs.umich.edu/filedispla y.php?wh ich=f&action=movie&id=66

The rest of the vids
http://kesisleme.eecs.umich.edu/media.php

Re:I've been slashdotted! YEAH! I'm so proud!

[Anonymous Coward]

I'd have just started blocking requests with a Slashdot referrer until things calmed down. If they aren't going to give you the courtesy of a "heads up", you shouldn't feel obligated to maintain availability for that additional traffic unless you have some spare time you want to put on it. Take whatever means necessary to keep your stuff up for the visitors that normally hit it, but if the Slashdot-induced load is going to bring it down, take whatever steps you need to to keep it up, even if it means rejecting Slashdot-referred requests.

Re:I've been slashdotted! YEAH! I'm so proud!

[Anonymous Coward]

I'm surprised how many people don't understand what a little advance notice can do. Even the Slashdot folks don't seem to understand this, or maybe they just don't care.

Most other major news outlets will give you a courtesy notification before sharing your URL in an article they're about to post.

So yah, there's lots of things site owners can do to mitigate the increased load Slashdot editors put on them. The site owners can temporarily allocate additional hardware or bandwidth to the site. They can mirror certain pieces of content on other sites, or even the entire site (providing an alternative URL for the Slashdot post). Or maybe if they can't do this, perhaps they can set up some throttling to allow "normal" requests to continue to come in while generally slowing or blocking requests from Slashdot.

Contrary to the belief of most Slashdot kiddies, most site owners want their content to be available. Not all of them can afford to do so at the levels of load a Slashdot posting can cause. Not all of them necessarily want to when they detect a Slashdotting at an inconvenient hour. Giving them sufficient notice to allow them to try and shore things up is better for the site owner, and usually better for the Slashdot user, who might otherwise not be able to get in at all.

It's just common sense and tact.

Has anyone made a joke about a RAID array yet?

[Trespass]

Just checking.

OMG MICRO$OFT SUCKS LOL

[Anonymous Coward]

hmmm

[TitanOfire]

If you eat it will it crawl back up your throat? I dont want those problems again.

Something that was really cool was....

[Anonymous Coward]

when they did this special about the brain and showed where this guy took the teeny brain of an eel and put inside of a robot, effectively turning it into a cyborg. They showed shots of it swimming around rather happily. I just wonder how the eel felt. "This is just bizarre! Every time I want to whip my tail around furiously, I move forward! And when I want to gulp seawater and pump my gills, I go around in circles. Oh my god! I'm a monster! What have they done to me? Oh!!!"

Comment removed

[account_deleted]

Comment removed based on user account deletion