PDA

View Full Version : Awesome exoskeleton from Japan, machining included



Evan
08-01-2011, 07:47 PM
Yes, it is a 21 meg video. Sorry for those on a slow connection but this is amazing. Made by students and they have the entire thing nailed.

http://ixian.ca/pics9/exo.jpg

video
http://youtu.be/IdPpWy_O09k

story
http://www.popsci.com/diy/article/2011-03/video-diy-human-powered-exoskeleton-makes-you-bigger-stronger-funnier


Add some power boost to this and the military will be all over it.

lazlo
08-01-2011, 08:12 PM
Pretty goofy, but this is what DARPA is spending our tax dollars on:

MILITARY SOLDIER EXOSKELETON
Defense Exoskeleton for bionic supersoldier

http://www.tacticalwarfightergear.com/tacticalgear/catalog/soldier_exoskeleton.php

http://www.tacticalwarfightergear.com/tacticalgear/catalog/images/defense_exoskeleton.jpg

Lockheed-Martin has a hydraulic exoskeleton, but the fatal flaw in all these designs is the power pack. Until you have a thermos-sized fusion reactor... :)

aboard_epsilon
08-01-2011, 08:12 PM
military ..how about total control and a police state in the not too distant future .

don't look so good then does it

all the best.markj

Black_Moons
08-01-2011, 09:17 PM
Pretty goofy, but this is what DARPA is spending our tax dollars on:
Lockheed-Martin has a hydraulic exoskeleton, but the fatal flaw in all these designs is the power pack. Until you have a thermos-sized fusion reactor... :)

I have a 20 pound 2 stroke motor that makes over 4000W (6HP) continiously (Well, Maybe a little more weight after we water cool it or add a proper 200W+ fan for stationary operation..) at only.. reasonable noise levels. Heh.

Then there is RC engines that can produce that in 2 pounds. of course, having to top your soldiers up with nitromethane might be intresting, And having them scream around at 30,000rpm...

PS: really hate those blue circles. Makes the guy look like an Iplod.

Weston Bye
08-01-2011, 09:35 PM
Power for the exoskeleton? Internal combustion is still the best for energy density, but...

Any air-breathing engine would be vulnerable to something as simple as a fire extinguisher. CO2 (or any inert gas) would deprive the engine of air, A dry powder cloud would react inside the combustion chamber, not to mention the abrasive/corrosive nature of the residue. Might even clog a well filtered engine.

A steampunk steamsuit might last a little longer after the fire went out. A chemical reaction steam drive like a torpedo engine would not be so vulnerable, but would have a short "sortie" time. Electric drive would similarly be limited.

Probably not a good choice for crowd control.

Evan
08-01-2011, 09:43 PM
There exist micro turbine power plants that fit in a back pack. I recall reading about one in the last couple of years that could easily power an exo like the one the Japanese students built. The primary use for suits like that wouldn't be actual fighting. It would provide the capability to cross even rugged country fast while carrying 2 or 3 times the usual load, most of that being ammunition and heavier weapons.

What the students have demonstrated is excellent control and balance with no active systems at all. That is a major achievement.

Black_Moons
08-01-2011, 09:58 PM
Any air-breathing Human would be vulnerable to something as simple as a fire extinguisher. CO2 (or any inert gas) would deprive the Human of air, A dry powder cloud would react inside the Lungs, not to mention the irrating/corrosive nature of the residue. Might even clog a well filtered gas mask.

Probably a good choice for crowd control is gases and powders

^- fixed that for you.

I wonder whats harder to filter and/or provide artifical air for. Humans or IC engines (sized for driving an exoskeleton, So lets say about 10x highly fit human power, or around 3hp..)

On the one hand, Humans need clean air cleaned of many toxins/dirt/debrie.
On the other hand, Its not that big of a deal if an engine dies, And they can suck of toxins no problem, Dirt and debrie is a bigger problem, But the whole engine could be replaceable, So if you REALLY had to, a few minutes in debrie/excessive dirt could be done.

Humans need little air compaired to an IC engine.
'Air' or oxygen can not be easily compressed, and can't liquify at less then very, very cold tempatures. Nitros oxide on the other hand, could be (I wonder if you can make an engine that can easily switch on the fly to just fuel and nitros oxide?)

Filling soldiers up with liquid oxygen (And nitromethane?) does sound like fun high tech ****! Sure would make great hollywood style explosions when someone gets hit.

IC engines can be restarted after oxygen is returned, or after the debrie storm stops. Humans.. Not so much.

So overall, Much more air needs filtering for an IC engine, but does not need to be filtered nearly as well then the air for its human pilot.
If your gonna provide them artifical air for CO2/other gas blackout conditions, It might be about balanced for runtime of nitros oxide verus compressed oxygen, given equal bottle sizes for each (Just a guess, no math done to back that up, it just seems like the factors would reasonabley balance)

Of course, being on a huge exosketon, it would also be much easyer to carry said large bottles of oxygen/nitros.

And while I have not heard of any area denial CO2 weapons, Nor soldiers running around with oxygen bottles, Nore dust weapons capable of cloging IC engine filters or dust masks... I must admit, Dumping a few 1000 liters of liquid CO2 as a weapon does sound intresting. Why must we allways blow stuff up? Can't we find anything else to try?

aostling
08-01-2011, 11:11 PM
What the students have demonstrated is excellent control and balance with no active systems at all. That is a major achievement.

Indeed, it looks like there is virtually no learning curve. Imagine one of these, fitted to your dog. Wouldn't he have fun?

Teenage_Machinist
08-01-2011, 11:33 PM
I think actually that helping the disabled is a major goal of these.

In my own (soon to be) neighborhood, some UC Berkeley people build a very inexpensive (i.e. costs 50K instead of hundreds of thousands) exoskeleton allowing a paraplegic Berkeley student to walk. He was able to walk at his graduation IIRC.

Evan
08-02-2011, 01:21 AM
I think actually that helping the disabled is a major goal of these.

Absolutely. That is how I happened across the article. I am planning an assist device for myself to make it easier to squat and get back up. Right now it is a major effort to get down to the floor and back up again. I can only do so a few times and then I need some sort of assist such as something to help pull myself up. I am thinking along the lines of a simple framework that straps to my legs and stores energy as I squat, probably in gas springs. When I need to get back up the gas springs provide the required assist.

I will have to do a lot of testing to find exactly what motions and degrees of freedom are required as well as the most comfortable way of attaching the braces to my body. An important requirement is that it not interfere with driving without having to remove it.

lugnut
08-02-2011, 01:28 AM
Damn Evan, when you get that thing figured out, you can come down and help me finish the laminate flooring job I started and couldn't finish. At least send me a copy of the plans so I can make one:D

Black_Moons
08-02-2011, 01:32 AM
I am thinking along the lines of a simple framework that straps to my legs and stores energy as I squat, probably in gas springs. When I need to get back up the gas springs provide the required assist.


Hmmm, I can't see how gas springs would work well. Maybe if small short ones placed off to the sides...

I would think clock springs on either side of the knee myself. Only problem is if your foot slips.. Ouch. Speed limited recoil is kinda needed.

Maybe something like a gas shock pulling on a cable, Wraped around a pully? (And the end tied through a hole in the pully, Since you only need 1/2 turn anyway), Then the speed limited retract would'nt get in your way either. You could disconnect the cables easily enough to completely disengage the rig... Or just have a length toggle mechanism...

Evan
08-02-2011, 01:56 AM
There are many factors to consider to make something that will work well. The knees are the easiest joint to assist because they have the most limited degrees of freedom but even so the mechanism will have to be able to accommodate some twisting and side bending forces. I don't see the assist system as being the difficult part. The geometry and attaching system will probably be the most difficult. It would be nice if the system could be somewhat locked in an almost sitting position. At least, right now that seems like a good idea. It may not be. A method of easily disengaging the assist system will probably be necessary, especially for driving.

I have checked with my doctor and the local pharmacy and as far as they know there isn't such a device on the market. When I build one, if it is successful, I will provide free plans for the system.

aostling
08-02-2011, 03:19 AM
In 1970 I invented what I thought was a novel spring motor. It turned out to be impractical, but for this project I derived the formula for the amount of energy (per unit volume) which can be stored in a material by stretching it. This equals one half the product of the tensile stress σ (measured in lb/in2) and the strain ε (measured as inches per inch).


Energy = σε/2

This will be in units of in-lb/in3.

Of all the common materials I found that the rubber has the best potential for storing energy by stretching. Here is how it compares with music wire:

music wire 4000 in-lb/in3
rubber 9000 in-lb/in3

Music wire has a much higher tensile stress (as high as 500,000 psi), compared to only 3,000 psi for rubber). But rubber has a far higher maximum strain (6 in/in) than music wire (0.016 in/in).

So you might consider using surgical rubber tubing for your energy storage material.

Evan
08-02-2011, 05:52 AM
Of all the common materials I found that the rubber has the best potential for storing energy by stretching.

A fact well known by those that build and fly model aircraft powered by elastic bands (as I used to do).

An elastic system is definitely a contender. Some sort of damping method will be required. Elastics have the considerable advantage of being cheap, easily replaceable and adjustable too. The material is widely available in the form of exercise bands.

taydin
08-02-2011, 06:06 AM
Of all the common materials I found that the rubber has the best potential for storing energy by stretching.

The redundancy provided by multiple rubber tubes is also quite beneficial. If one of them ruptures, the rest will still keep the tension intact. You'll only feel a slight increase in pressure. Otherwise, a failure in a single spring arrangement could suddenly put your entire weight on your knee joint.

Yow Ling
08-02-2011, 06:19 AM
Power it with Warsop hammers that John posted a couple of weeks ago
Would only take a few minutes to get the knack of running 2 at once

http://www.youtube.com/watch?v=XnnvH5_Y0Eo

aostling
08-02-2011, 01:21 PM
An elastic system is definitely a contender.

An elastomer can store about 9,000 in-lb per cubic inch if used in tension (or about half that much if used in torsion). Your proposed task will require storing about 500 ft-lb of energy (250 lb for man and machine, raised vertically 2 feet), or 6,000 in-lb.

So you will need to incorporate at least 2/3 cubic inch of (say) rubber tubing into your design.

And probably significantly more. The estimate of 2/3 cubic inch assumes you stretch it six times it's relaxed length, as rubber is capable of doing. The design need not do this of course, but the amount of rubber required will double if the strain is 3 instead of 6.

Alan Smith
08-02-2011, 01:45 PM
Isn't there supposed to be a very significant creep factor with elastic/rubber which results in huge losses of stored energy unless the energy is released immediately. I think there was ballistic work done using surgical latex tubing in catapults that showed that even a few seconds delay in releasing a shot resulted in significant fall in "muzzle" energy. This would make it a poor contender if you planned to stay on your knees for any length of time.

aostling
08-02-2011, 02:03 PM
Isn't there supposed to be a very significant creep factor with elastic/rubber which results in huge losses of stored energy unless the energy is released immediately.

I was thinking Evan's machine would be an assist for some creep-free tasks, like this: http://www.youtube.com/watch?v=gEYiioT-fjo

Evan
08-02-2011, 02:50 PM
I am more interested in being able to get down to floor level to pick things up. My knees aren't the actual problem. My muscles become severely fatigued after only a few repetitions of any form of heavy load use. I also don't need a full assist. Even taking off half the load would be sufficient and a variation in time of the amount of assist wouldn't make much difference as long as there would be a significant assist. It would also help with climbing stairs. The human musculature has very little ability to store and release energy. It's a very lossy system in that respect.

EddyCurr
08-02-2011, 03:49 PM
While untethered exoskeletons face delays to market because of power-
vs-energy consumption trade-offs, there are many applications where
tethered suits would be perfectly adequate today.

Put anyone of modest stature in such a suit and watch for productivity
gains, whether handling packages on a loading dock, components on an
assembly line, or ordnance on a flight deck. A genset would enable such
tethered suits to be used further afield.


Raytheon/SARCOS Exoskeleton (http://www.youtube.com/watch?v=0hkCcoenLW4)

Raytheon/SARCOS Exoskeleton 2.0 (http://www.youtube.com/watch?v=mO0xNI3xpmE)

I thought that bulkier versions of such suits had already been in service
for some time. However, a quick search does not support this.

Exosuit cargo-loaders appeared in the films Alien ('79) and Dave ('93)
but apparently these were strictly non-functional props,

.

lazlo
08-02-2011, 04:31 PM
[indent]Raytheon/SARCOS Exoskeleton (http://www.youtube.com/watch?v=0hkCcoenLW4)

Ah, that's the other exoskeleton I mentioned in my first post -- Raytheon, not Lockheed. They were plugging that video like crazy when IronMan 1 was released.

Cracks me up that they have to have a "test pilot" with a crash helmet. Wonder how many times he got bent backwards? :)
The crew of lab-coats in the background carrying the hydraulic lines are amusing - modern version of Steam Punk ;)

Black_Moons
08-02-2011, 04:44 PM
While untethered exoskeletons face delays to market because of power-
vs-energy consumption trade-offs, there are many applications where
tethered suits would be perfectly adequate today.

Put anyone of modest stature in such a suit and watch for productivity
gains, whether handling packages on a loading dock, components on an
assembly line, or ordnance on a flight deck. A genset would enable such
tethered suits to be used further afield.

.

A genset can be built small enough to fit onto a decent sized suit.
It might be a little noisy, And run at a non standard 5000rpm+ producing DC via alternator, Or just running a hydraulic pump and minimal control electronics.

Or 10,000 (gearbox) to 200,000rpm (direct drive?) for a turbine... Small noise issue with turbines of course.

Considering military aircraft have maintence scheduled in hours, not days of operation, I would'nt think that a exoskeleton suit that required a rebuilt motor every few 100 hours would be a huge issue, And muffler systems are not exactly heavy, just bulky. Mass produced 2 and 4 stroke 6HP~ engines can be gotten for under $150 shiped.. In quanity 1.

Random though: I wonder if one could use the exoskeleton frame as part of the exhaust system? Large hollow pipes are great structural elements, And exhaust pipes/muffler elements.

Consider how quiet a REALLY well built motorcycle exhaust system is. Iv heard some well under talking volume while producing power.

Also once seen a (very old honda? iirc?) generator, normal looking. 4KW maybe?.. but while under load, I was standing RIGHT next to it not 2' away, having a conversation with someone in my normal talking voice, No problem at all.

Evan
08-02-2011, 05:14 PM
Power won't be a problem. Here is an article that is directly on point and illustrates a 100 watt turbine power plant that fits in the palm of your hand.

http://ftp.rta.nato.int/public/PubFullText/RTO/EN/RTO-EN-AVT-131/EN-AVT-131-10.pdf

EddyCurr
08-02-2011, 05:21 PM
A genset can be built small enough to fit onto a decent sized suit.I might not have been clear that I meant a 'conventional' hand or
trailer mounted genset, positioned remote from a tethered ES.


It might be a little noisy, And run at a non standard 5000rpm+ producing DC
via alternator, Or just running a hydraulic pump and minimal control electronics.The BigDog robot from Boston Dynamics (http://www.youtube.com/watch?v=b2bExqhhWRI) could serve an example for a
self-powered exoskeleton (though it wasn't very stealthy at the time
of this video.)

.

aostling
08-02-2011, 09:20 PM
Here is a possible configuration for a "squat-assisting" mechanism. It uses an elastic cord in tension to store and release energy during deployment and release.

The large links would strap to the side of the user's upper and lower leg, with the pulley at the knee. Squatting extends the cord – as shown here it doubles in length (a strain value of 2).

The strain could be increased to a maximum of 5 or 6 by locating the cord anchor point closer to the knee pulley. Experiments might determine which strain level, or "leverage," feels best.


http://i168.photobucket.com/albums/u183/aostling/Screenshot2011-08-02at55234PM.png

Black_Moons
08-02-2011, 09:28 PM
Here is a possible configuration for a "squat-assisting" mechanism. It uses an elastic cord in tension to store and release energy during deployment and release.


I think one of the requirements is it not interfear with siting down on a normal seat.

Evan
08-02-2011, 11:29 PM
That's right. The mechanism needs to stay pretty close to the same envelope as the body. One way to accomplish that would be a spur gear at the knee pivot point that would drive a small pinon. That reels in an elastic or spring to stretch it.

aostling
08-06-2011, 10:16 PM
I think one of the requirements is it not interfere with siting down on a normal seat.


That's right. The mechanism needs to stay pretty close to the same envelope as the body.

As I was hiking on the mountain today I thought about another requirement: the spring should be disengaged when walking or sitting. If it is not, walking would not be as efficient, since you would always be expending effort to stress the spring, or to control the release. It's the same with sitting – you would not want your bent legs to straighten out while you are relaxing in that posture.

How might this be accomplished? I'm thinking the mechanism should have a lever which can be thrown by hand, an over-center mechanism which pulls slack elastic cords into tension. So if you were sitting down, with the cords slack, you throw the lever by pushing out and down with your arms.



Without regard to this requirement, I did some calculations on the feasibility of using a steel spring, either a spiral coiled spring like the mainspring in a clock, or a cylindrical helical spring like on a mousetrap. Either configuration stresses the material in almost pure bending, and Mark's Mechanical Engineers' Handbook has a formula for the volume of material required, depending on the required energy storage.

For a spiral coiled spring

V = 6EU/S**2


E = modulus of elasticity = 30,000,000 lb/in2 for steel
S = safe working stress = 150,000 lb/in2 for high tensile steel
U = energy stored = 1200 in-lb (for 50% boost of a 100 lb torso squatting down 2 ft)

Then V = 9.6 in3. That's about 2.8 lb of spring steel, 1.4 lb in each leg.

So a steel spring meets the energy storage requirement (just barely) without being too heavy. An elastic cord might be easier to incorporate into a disengagement mechanism, though.

Evan
08-07-2011, 12:13 AM
That amount of steel is much too heavy to be comfortable. I have a good idea of what I want to do now and when I get time for it I will do some drawings. Already this thread has been worth while since the idea I have came as a result of this discussion. And, yes, it must be disengaged for sitting and walking. I also have a simple way to do that too, simple enough that it is nearly instant like flicking a switch.

Smokedaddy
08-07-2011, 12:22 AM
Something else to look at (design wise) ...

http://www.mech.utah.edu/ergo/pages/Research/a5/index.htm

http://www.springzback.com/

-SD: