Slashdot Mirror
Homebrewed Robot Exoskeleton In Alaska
museumpeace writes "CNET has an article about a robotic exoskeleton ginned up by tinkerer from Alaska There are a few cool pictures. The audacity of Mr. Owens project, if you believe the article, compares to the efforts of the old
Home Brew Computer Club when compared to the work of GE or
Toyota. Inspiration here comes more from sci-fi and video games than from industrial competition. The article is a good roundup of MECHA related developments, some of which sprang from DARPA money, so I am glad at least a few of my tax dollars are having some real geek fun."
I thought they found an ancient robot skeleton buried under the ice or something, it's been a long week.
Well I for one, welcome our new Alaskan Mecha overlords.
If brevity is the soul of wit, then how does one explain Twitter?
Y'know, it occurs to me that this is someone you don't want to piss off with a slashdotting. He's developing the perfect retaliation suit!
John
But, I have a feeling the first step is going to put this flat on it's face.
Anime-ish designs for Mecha (mobile suits) don't translate well into the real world.
If it were me, I would be looking to emulate some of the designs from the original MechWarrior series. They seem a bit more realistic.
Still, love to see this walk/work!
DAMN YOU OCTODOG! DAMN YOU TO HELL!
HAAHA!!!! I guess not. Good, I'd like to think I have some kind of life. Seriously though, I usually find, that alot of the stuff I read here, previously on Fark.com.I imagine, there are a finite amount of sources for good material,though.
My cat's picked up a Hammer. HEY! Put down that Hammer. Put Down that Hamm...THUNK!
http://www.nfb.ca/grizzly/suit.html
Now THAT would be cool.
it gets lonely in Alaska. :-)
Rhymes that keep their secrets will unfold behind the clouds.There upon the rainbow is the answer to a neverending story
I hereby demand a battle royale to the death between the homebrewed robot exoskelton and this guy.
learn, when, to, use, commas,,,,
Cute addition there with the head crests a la "Patlabor." And yet I'm still rather disappointed. Where is the 14' energy sword? Jet boosters? Particle beam rifle? Facetious, yes, but it is strange that amid these dizzying technological advancements, humanity's achievements in the field of robotics, circa 2004, are analagous to the state of automotive technology circa 1904. Nevertheless, that is some pretty damn fine backyard engineering.
"OH SHIT, THERE'S A HORSE IN THE HOSPITAL!"
Oh great, put the robots in the place where we can't live. They'll multiply... then they'll develope new, improved A.I. They will take over the economy with their new hoverdrive technology... We won't know if it is us or them who'll attack first-- I just know we're gonna have to scortch the sky because of this.
Electrons are free; it is moving them that becomes expensive.
I once caught a fish this big......
obscure movie reference...
__________ Leave me alone I'm compiling a RPG II program on my S/36...Thanks to metamucil I'm a Regular Meta Moderator
The control problems are not trivial, and I doubt that this guy will be able to solve them. But a lot of these problems have already been solved by Sarcos. In particular, look at this page, especially at the "Sensuit" and the "Large Arm."
The Large Arm is especially impressive, holding a freakin anvil like it was a stein of beer!
If you could build the whole body of the Sensuit to a large enough scale that the whole pilot can be encased in a haptic feedback harness, you'd have a viable mecha.
The reason no one has ever actively pursued mecha is because they would be so inefficient. Tanks and such are built with a low profile, and if you ask frontline soldiers, they are only useful in limited roles. My army friend much prefers the new strykers because of their flexibility and reliability.
A mecha would be standing 20 feet in the battlefied, an open and tempting target to everything from bombers to tanks to helicopters and to RPGs. It would have limited mobility, be extremely difficult to keep in working condition, and will have less load capacity than its tracked or wheeled counterparts. In short, it would look cool, but would be a useless coffin.
In BattleTech, they make up for the obvious disadvantages of a mecha by giving them advantages over vehicles. Mecha are more reliable, more maneuvarable, able to take more damage and continue to function, and can carry more weapons. Even then, if you pit a balanced vehicle force against a balanced mecha force, ton for ton, credit for credit, the vehicles can easily overpower the mecha in most circumstances.
I don't want to discourage this backyard project. After all, how many inventions were made when there was no necessity, but a necessity was found at a later time? But I do want folks to exercise a bit of common sense. If mecha were such a great idea, we'd have used them in WWII. We certainly had the technology to build them back then.
The radical sect of Islam would either see you dead or "reverted" to Islam.
To build an operational mecha as a hobby is one thing.
To build it outside in the friggin Alaskan winter... well, there's only one word for that:
OTAKU!!!
(which, by the way, is what is on my housemate's vanity plate, so I speak with first-hand knowlege)
Someone had to do it.
Although I do agree that it'll probably fall flat on its first step. Cool backyard project though.
My dream of finally being one of the Robot Jox might soon be a reality. Sweet!
This is the beginning I tell you. We must destroy SkyNet while we still can!
He's paid for the entire thing himself (about $15,000). DARPA has funded other powered armor projects, but not this one.
Mod down posts with a "Free Mac Mini/iPod" sig, they're spam!
Ah, the irony there would be...
that that is is that that is not is not
No, the Darpa project is completely seperate. Darpa is sponsoring a Berkeley project which already has working legs and can carry oversized packs. For an image of the Darpa-funded, Berkeley project go to here
The things men do when there are not enough women around. I guess it keeps his hands from going crazy on those long alaskan nights.
"How perfectly Goddamn delightful it all is, to be sure" Charles Crumb
Wow someone used the word "control". After working on a VERY, VERY... VERY simple servo motor control project I have to agree with the skepticism. Who knows though I am stupid, other people aren't. I also haven't dedicated any "real" portion of my life to control systems. Also someone else brought up the impracticality, which seems to be a logical arguement to me. But again I don't really know that much. Overall it is pretty cool for what it's worth!
It takes all types in this world. I sincerely mean it... This is just my perspective.
It is obvious that this was not designed by rogue chicken scientists. It is nowhere near sleak enough.
Forget this guy in the post. He clearly doesn't have a clue. But the problems have been largely solved in the past several decades with DARPA money.
If you put a full body haptic interface around someone strapped into a huge robotic body, you'd have it. (See the Immersion Corp link.) But the thing would have to be freaking huge. A full-body haptic cockpit would be something like a sphere 8' in diameter, implying a mecha 30 foot tall!
Perhaps have the cockpit controlling separate and much smaller mecha body remotely, and just have the haptic controls on one of those motion simulation platforms.
Sure, it wouldn't be as "sexy" as the X-Prize, but wouldn't some privately sponsored prize money do wonders for this longtime human dream? Call it, say, the Mecha Prize, and offer a few million bucks to whoever builds the first mecha that can go a half mile, pick up a Dodge Neon and move it in the air for 50 feet, then return to the starting line. Or something similar.
I have no doubt that someone as creative as Rutan is out there, and with a little incentive and the promise of some real financial gain could use modern actuators and pressure pads and gyro sensors and so forth to finally create a useful mecha.
I also have no doubt that (unlike Spaceship One) a mecha that could complete the above test would immediately be of great value in quite a few industrial and/or emergency applications.
--Brandon / Split Infinity Music
Mechas make perfect forklifts and industrial machines. There's some use for combat mechas in exotic terrain... mountains, underwater, etc
:)
They'd make excellent car crushers, too
Honey!, Where did you put my Super Suit !?!
Let's hope his server has a hefty exoskeleton as well...
it's fark without the boobies
It's hard to believe that's how Micronians are made. Why don't we see it right now by having you both kiss one another?
"...and a gas engine mounted on the back to generate the power needed."
i hope he's got intentions to incorporate some sorta personal heating system....-40 and metal is not a good combination
This sig contains repetition and redundancy.
Wait til they team up with the Tron Guy.
Table-ized A.I.
From the article:
Yeah, I'd hate to be negative (and this is a totally cool project, kudos to the guy) but that's the first thing I thought: getting this thing to walk around is going to be a challenge.
... still, can you imagine the complexity of staying upright while you lift and lug heavy things around? Heck, even leaning against something to push it while on two robotic legs is complicated, much less ripping up trees by the roots and swinging them over your head Anime-style (I exaggerate, but this more or less seems to be the goal).
However, he's not trying to build an automonous robot like Honda or Sony. To me it seems that an exoskeleton is the place to start: if it wasn't for the fact that walking upright on two legs is a bit of a controls hat-trick!
I'd imagine that a robot on wheels, with arms that move as you move your own arms would be a more attainable first step for a hobby.
Then again, maybe this guy really does have some clever ideas as to how to walk around without tipping over
It'd make a great extra-planetary exploration device if it were sealed. Could you imagine these things climbing around Valles Marineris where the scale of the enviroment would match the scale of the suit.
Seems like I know this guy from somewhere, but can't quite place where....
http://www.paralinks.net/paralinksarchives/hawking exo.html
You need a mecha in Alaska to do all that tongue-touching metal work during the long Alaskan winters.
People get out of Gitmo faster than people get freed from touching their tongue to metal in perma-front country.
Why people insist in calling "robot" something that is not a robot?? A robot must be autonomous , that is, no human intervention involved.
Anyone remember or know what I'm talking about? I can't seem to find anything about it.
Take a look at my Other Post for some links relating to the research that led up to the accomplishments of Sarcos. Not flashy stuff. Mostly concepts from old research. But if you put the concepts together, you realize that Mecha are indeed possible. Though I would agree with the other guy that pointed out that these are impractical for real warfare. But I would still watch a TV show with 2 human piloted mecha fighting!
Am I the only one who sees an 18 or 35 foot tall mech being build surrounded by scaffolding and thinks that the whole project would have been 75% easier if he judt built it lying down?
A crane could easily stand it up later, but if it is going to work at all, it would need to be able to stand after a stumble anyway.
Sure, it looks cooler standing, and probably annoys his neighbours more, but it seems like a very poor design decision.
Big difference: he's using mechanical linkages and hydraulics to operate it--not electronically controlled motors.
Have you ever watched a backhoe operator working? Or, for that matter, driven a car? It's a matter of relying on muscle memory to adjust for slop in the mechanism.
I don't think his project is much different than other machines on the monster-truck circuit (read center paragraph for a riveting description of two giant battling robots). This guy's design is most notable in that it departs from the traditional transforming metal dino-sar theme.
The point is it's not rocket science to design a "control system" for what are basically jerry rigged front-end loaders. In fact, it's demolition derby science: rocket science's archnemesis.
Granted, it would be a bit more tricky to get it to actually walk, but I imagine the operators will just wheel the thing out on a flatbed and wave it's arms around a bit and shoot flames. All his talk of military and forest fighting capabilities is just successful buzz to get rednecks like me out at the local speedway.
I'd say that depends entirely on what human features one chooses to enhance. Think of the possibilities!
-scsg
This is way cool and it'd be awesome to see it work, but I'm officially a doubter.
Tanks are fine for semi-open country, at least when nobody's throwing tactical nukes or shooting with depleted uranium shells, but they have some vulnerabilities in cities. During the 1956 Hungarian Revolution, the Russians were invading Budapest, and the locals would take out a strip of cobblestone street, pour gasoline in it, and light in when a tank crossed, with negative results to the tank. Probably newer tanks are a bit better protected, but it's still a risk.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
Korea is not the country with the mecha-craze. It is Japan. The building of a giant robot is not new. Many years ago, this giant robot was already being developed in Japan. Although there is no actual functionality with this Gundam, but I doubt that NMX04-1A can be made to be autonomous any time soon.
http://www.paralinks.net/paralinksarchives/hawking exo.html
g exo.html
/. does that to a line of text, but the URL thing doesn't add spaces [to the ACTUAL LINK - the text still shows a space in hawkingexo].
Actual, operational link:
http://www.paralinks.net/paralinksarchives/hawkin
Please use the URL feature that will auto-link a URL, as shown under the Submit button on the Post Comment page. Those of us who perfer to click rather than copy-paste and then fix the link.
In your post it has a space that's not in the actual link, dunno why
Tag lost or not installed.
On this page there are photos of exosuits which were at Robonexus in October in a mecha lifting contest.
Mechanicus is another home-made exoskeleton from Austin, TX.
Anyone ever hear anbout the Tetsujin contest? "Teams build powered exoskeletons to augment the strength of one human in a weight lifting competition." Sounds like fun! I can't wait until more people build these things. Then, as the natural progression of robotic projects seem to always head to, the next generation of robot combat can take place! Think 'RobotJocks'... :/
http://www.servomagazine.com/tetsujin2004/
he's gonna try the modded Xbox and Steel Battalion controller. Wish him luck.
If /. was imitating FARK, the article would have been described as "Man designs mecha; will crush cars, find Sarah Connor."
for which he won an Ignoble Prize. http://www.improb.com/ig/ig-pastwinners.html#ig199 8
Ignorance is curable, stupid is forever.
...you lose your turn.
*whup* "Get along, little electrons. Heeyah!"
Actually, the Sarcos Large Arm uses hydraulics. In fact, only hydraulics are powerful and responsive enough for machinery like this.
Real science is better than demolition derby science any day. Instead of a couple of humanoid megasaurs lumbering around, imagine two giant robots nimble enough to do Kung-Fu! (Duking it out with giant clubs!)
I goth my thung thuck thoo the thoot.
See I told you, pr0n drives all technological advances! "Exoskeletons for Human Performance Augmentation" is just a fancy name for a strap-on.
Well you can't justify a few millions on research for a "GIRL ROBOT" which if fully integrated with all sorts of pleasuring features.
You can however, justify it when you go for "humanoïde robots" and add funky features later.
You don't think those guys inventing the internet were actually building a "network to withstand a nuclear attack"? : P
Just be patient my friend. ^_^I think we can keep recursing like this until someone returns 1
Most of the feedback doesn't need to get back to the brains of the unit; instead it can be handled by computational processors. Consider the way that Kamen's iBot and Segway prevent most of the feedback from going to the pilot. In the case of the iBot, 3 Pentiums do the calculations to keep the machine responding in fashion appropriate for the pilot's intent.
Is that thing supposed to be manned? Someone make sure to find out if he survives when he tries to make that thing walk...
Sleep is futile.
Ah!
This was one of my favorite shows in the 80s.
I saw the French Canadian Version!
It had songs and lyrics!
Much better than the american version I believe.
Here is a link for nostalgia.
FULGURO-POING!
http://www.vex.net/~guru/goldorak/goldorak.htm
Now, all I want to know is: when can I order one for myself? Not that I have any earthly use for such a device, but it's just too damn cool!
Doing my level best to piss off the religious right wing...
The problem with backhoes is thery are:
1) Worked so hard and long (due to fiancial pressures) that their joins develop slop
2) The hydraulic valves only have a limited amount of proportional control
With a well maintained machine and good proportinal control valves, hydraulic force can be applied very smoothly and controllably. The power/size radio is incredible and is really required for something like this.
What if the guy GPLed the thing?
/. anyway so someone had to bring this up. BTW I see the meme war has already started so let me add:
What would military do?
Osama (just to use some wildcard name for the evil character they want you to fear now) could use it as well as caaptain america (insert your favourite hiro/protagonist [blink])
Would anyone acknowledge the importance of open developing and GPL derivative work or would they just secret it up?
Ok I really have no idea how open source licences would apply in such a scenario but it could be interesting
[and this is
now imagine a Beowulf army of theese!]
I can understand this, but the challenges remain. In order to balance like a human being, the mecha will have to have a lot of the same flexibility and dexterity -- not impossible but incredibly hard.
....
1 _2 _3_0_5.html
The huge robotic legs (and arms!) will have to be incredibly sensitive and effective to pick up the subtle movements required to even just stand in place. I'm sure he's going to have to figure out how to use proportional control. That is, I think he needs to move a limb slightly and slowly or fast and with more force in order to balance and move like a human. I'm not sure how you'd do this without sensors and computers providing control of motors
Basically, the input required to move a massive and heavy robot arm is different than the movement we might make with a human arm. Since it's more massive / dense / heavy / powerful than a human arm, complicated inputs (reverse power before you need to stop a movement so you don't overshoot your desired position, yet complete movement ASAP) are required for any movement. This analysis is the sort of thing we do subconsciously with our muscles when we move our own limbs, but isn't reflected in the movement itself. Nor is the "controls" that is used by my brain and human arm the same that would be required for an arm with different dynamics.
So if I rigged a big robotic arm to follow my movements it would have to have computer intelligence that takes into account the dynamics (inertia, degrees of freedom, etc.) of the arm in order to achieve the desired movement. Without computers, the arm will most likely wail around wildly as it tries to match my movement (see below on bang-bang versus proportional control -- well written IMHO). Bang-bang controls are doable mechanically (as simple as switches that trigger movement), and I can only think of one example of proportional control done without computers (James Watt and his flyball governor). I'm sure there are others but purely mechanical proportional control is the exception (and harder to do than to just use a computer).
I don't understand how the movement will work if he sticks to not using computers for this reason. "No computer on huge heavy powerful limbs" equals "hard to control movement" equals "hard to walk!"
It is entirely possible that I just don't get it; I just think that a lot of the difficulties in making a robot walk this way are easy to understate since we walk without having to think about everything that has to go on for us to balance on our two legs, or even just move our limbs.
Don't get me wrong, ever since Robotech I share this dream, and I am very curious to be proven wrong. Hell, I'd be one of the first in line to order one from him it works (well, I'd have to win the lotto first). He does seem convinced that it's going to work out: he's already got plans to sell prototypes when the bugs are all ironed out.
I found this link on controls that mentions at the bottom
Proportional control makes systems run smoother than bang-bang. Most biological systems are proportional, while many engineered and all too many political and social systems are bang-bang
http://www.fourmilab.ch/hackdiet/www/subsection
Admittedly they have done a lot more theoretical and design work than just slapping angle together then making it move, but http://www.mechaps.com/ have a very clever project ongoing. Check out the images section for up-to-date concept and design sketches.
Y'know... shadow moses island was supposedly in alaska....
Peace, Robot Monster. I just.... wanna....dance...!