Japan making power exosuits for real...

[Manedwolf]

Awesome. Negates its own weight, and reduces the apparent weight of an object you pick up by 50%.



http://news.yahoo.com/s/afp/20071006/tc_afp/lifestylehealthtechnologyjapan_071006174844

Except they're still aiming for the wrong market. They need to look at military applications.

The Army and Marines want to meet their recruiting goals? Get Japan to develop hardsuits. Who wouldn't want to wear an armored, powered, bombproof, small-arms-proof exosuit that let you tear doors from hinges and pick up cars? They'd just need to develop the "BFG"-class of rifle to go along with it. 

[Kyle]

Ok, thats friggin sweet.

I have this image in my head of a new imperial japan...

"I know our constitution restricts how large of an army we have... And in truth, we only have 37 soldiers in our army. BUT THEY ALL FLY GUNDAMS! BWAHAHAH!"

[LadySmith]

Kyle, you owe me a keyboard. 

[Kyle]

This is no laughing matter.

What are we gonna do when the japs find their gundamite reserves?

[DustinD]

I know gundams are really powerful, but if we could defeat their tanks in WWII despite the fact that when five of them got together they could form into giant robots we should be ok now.

[AJ Dual]

Exoskeletal armor? Believe me, DARPA and it's ilk are working on it.

Japan has sort of a top-down approach to robotics in general. They tend to just try and make "a robot& any robot" then refine it refine it and refine it& If you're familiar with Honda's "Asimo" and all it's precursors you're familiar with the approach. It's sort of a "concept car" approach to robotics. Quite often, those "amazing Japanese robot" videos are just very expensive puppets making 100% pre-choreographed movements.

http://www.youtube.com/watch?v=ASoCJTYgYB0

Did Asimo's vision system scan the steps, and then calculate their size and position incorrectly? Or were the stairs simply an inch over from where the programmers scripting Asimo's movements were told they would be? I know Asimo has to have some sophisticated balancing and position control to walk bipedaly, which was a challenge Honda engineers were working on for years, but I also know Asimo's "brain" is running over WiFi, and it's not making it's decisions locally etither.

Here in the States, DARPA here, or a university there is working on all sorts of robotic or exoskeletal problems, very often our prototypes look junky and unpolished, but more often than not, the underlying technology is very, very advanced.  There's very heavy emphasis on instilling "reflexes" neural networks that replicate and uncover the reflexes that animal life has perfected over millions of years. Autonomous semi-intelligent limbs that don't need a big "central brain" telling them what to do, just like your own feet and knees need your brain telling them what to do every second so you don't fall over.

Watch this DARPA "packbot" making un-choreographed movements outdoors in unpaved environments.. http://www.youtube.com/watch?v=x3G-UE1HwGI take special note at about 0:30 sec where the man walks up and kicks the packbot, and it "stumbles" and regains it's balance. It's eerily like a "real" quadraped animal when it does so. The software and microcontrollers are making those decisions on the fly. Often there even isn't a "central" program at all, just nested series of reflexes that function in a hierarchy. Quite possibly giving insight to how insects are capable of doing so much with so little in the way of neural matter.

And the DARPA grand challenge, the Stanford team won with a modified VW Toureg SUV, which independantly navigated 132 miles of Nevada desert, a mixed courseway of open terrain, and roads, it made it's own decisions and detected all obstacles with it's own scanners and vision systems. http://www.grandchallenge.org/

I'll go out on a limb and say that we all know someone in our circle of friends and family who is a licensed driver, but probably wouldn't have made it&

I think exoskeletal powered armor for infantry is quite possibly an excellent idea. However, we'll probably never see multi-story "Gundams". Even with what will be possible with futuristic materials science, the same engineering problems that limited the maximum attainable size of the dinosaurs will come into play. And perhaps more importantly, a TALL bipedal walking robot will be one primary thing on the battlefield.

A bullet magnet.

A HMMV and a TOW missile could easily take out such a robot from a safe distance, how much did the HMMV and the TOW cost? How much did that Gundam cost? Now, a exoskeleton that could duck behind walls, jump over them, or hide in a building AND fire it's own TOW back at that HMMV, that's another story...

[Werewolf]

A HMMV and a TOW missile could easily take out such a robot from a safe distance, how much did the HMMV and the TOW cost? How much did that Gundam cost? Now, a exoskeleton that could duck behind walls, jump over them, or hide in a building AND fire it's own TOW back at that HMMV, that's another story...

Which is exactly why - as way fun as they are to play - the whole Mechwarrior series of games takes a major suspension of disbelief to play. I mean - come on - 2 to 3 story tall machines, walking on two feet and piloted by a human yet missiles are barely guided, aren't fire and forget and have ranges of only 2 or 3 KM's. And in a society that's capable of interstellar travel...  - puh-lease.

Still - the whole concept of man-sized combat armor vis a vis Starship Troopers (though not nearly as advanced) does seem like a real possibility. Maybe in the next 20 years we'll see something like it.

As one poster postulated:

The Army and Marines want to meet their recruiting goals? Get Japan to develop hardsuits. Who wouldn't want to wear an armored, powered, bombproof, small-arms-proof exosuit that let you tear doors from hinges and pick up cars?

If we had it and no one else did? Imagine the possibilities.

[AJ Dual]

Yes, I was thinking of Mechwarrior too.

I can think of an even bigger reason to need exoskeletal armor in the near future.

Look at the work that's being done with "swarm bots". Robot's who's individual intelligence is quite limited, but they have rules for interacting and producing emergent group behaviors to accomplish tasks.

Just like ants.

Now imagine thousands, millions of these deployed on a battlefield, about the size of cats, maybe rats. Each of them has a 1/4lb of C4 in it's belly, and it's programmed to seek out humans and kamikaze. The ammonia in sweat, voice/speech recognition of the enemy's language, image recognition of the enemy's uniforms and weapons, or simply just things that emit IR at 98.6 degrees.

And some factory in Chinostan is turning them out for $19.95/ea in bulk. And there's flyer models, spider models, wheeled jobs like those Segway scooters, really little things that roll around like hockey pucks or tennis balls. They're deployed from the air by UAV's in cannisters like the "old-time" cluster muntions, or scattered behind as the enemy "retreats" from an area.

In essence, imagine a minefield. But the mines are robotic ants, roaches, birds, and rats. And instead of passively waiting to be stepped on, that minfield is talking amongst itself and is hunting for you. All coordinated over wi-fi. The little "spider bots" that swarmed the ghetto in "Minority Report" looking for Tom Cruise, retinal-scanning everyone is kind of close, but imagine 100 X of them, and they're disposable anti-personell devices.

Then there's bigger ones that seek out armor and vehicles, and kill them with self-forging copper warhead disks. And full size UCAV's, with all the technology of F-22's, but no G-limits on it's manuvering envelope, having done away with that pesky fragile human in the cockpit...

Of course there will be swarms of anti-robot-robots too...

[Joe Demko]

The exo-armor thing has a science fiction attraction about it, but I'm not convinced of its real world utility.  In many ways, it shares the same flaws as the genetically-engineered supersoldier plan.  How strong does a soldier have to be?  Essentially strong enough to carry and use his weapon/ammo/equipment load.  Which is going to be more cost-effective and allow a greater number of soldiers to take the battlefield:
1.  Powered armor suits with integral weapons systems.
2.  Lighter conventional body armor and lighter/smarter weapons.

Cool as they are, gundam-type things are basically just another AFV; a light tank with legs instead of treads carrying one guy instead of a crew.  An anti-tank weapon that will take out a present-day AFV would almost certainly obliterate a powersuit.  Further, can a suit carry enough fuel, munitions, and ammo (plus necessary water and rations for the pilot) for any kind of extended operation?  I can easily see these things being so maintenance-intensive that their supply train would be enomous.
Kewlth aside, my big question is what will powersuits do that makes them worth the crazy amount of R&D expenditure?  I'm willing to be convinced, but I see them as not much more than tomorrow's version of an armored car.

[Manedwolf]

I also can't see a reason for giant robots, as they would indeed be taken out by low tanks hugging the terrain, or shoulder-fired weapons...or an over-the-horizon shot from an aircraft.

I do, however, see a major reason for personal armor, especially if we're involved in the current theater for the long term.

Suicide bombers.

I just saw an article wherein an insurgent was saying how he had his brother go from a drinks stand with Pepsis in hand, walk over to some American GI's with a smile, offering them the drinks...and then blew himself up beside them.

With a powered, shrapnel-proof exosuit with dampening filters on incoming audio, such an event could just result in a need to be hosed off, not a medevac and lost limbs or bodybags.

CQB in powered, compact armor that was impervious to small-arms fire and booby traps, with sound enhancement, thermal imaging and a "moving map display" of the building and where your squad members are would be a big help to soldiers, I would think. (especially if it kept them comfortably cool as well!) It'd also have a severe intimidation factor to enemies.

[Joe Demko]

Barring some really revolutionary breakthroughs in armor, powerplants, and energy sources, I don't foresee powered armor that ends up being very small.  It has to be big enough to enclose the pilot for just for starters.  Start adding armor and it's going to start getting a lot bigger.  Start adding weapons and ammo pods and it gets bigger yet.   By the time you're done, will you have something that is notably better than a modernized version of the early "tankettes?"  What will making it anthropomorphic offer over just making an armored all-terrain pod?  If you end up with a power suit that is 10 feet tall and several feet wide and thick, there doesn't seem to be much of an upside to wearing it instead of riding in a tankette the size of a Cooper Mini.  The power source is going to be the stumbling block for these things.  Short of perfecting a "pocket" fusion generator, I don't see them ever coming to fruition.  The prototypes, to the best of my knowledge, are externally powered.

[BrokenPaw]

Joe, the one possible reason why a powered armor suit might be better than a similarly-equipped "tankette" is the user interface. 

Powered armor along the lines of that in Heinlein's Starship Troopers is worn, not driven.  The cap troopers didn't have to pilot their suits, they just moved as if they were going about their business, and the suits magnified their movements in force and range.  The soldier who doesn't have to turn a steering wheel, but can just turn naturally to face a threat while using strength-enhanced hands to bring weapons to bear arguably has a tactical advantage over the guy driving a tankette and using a less-natural interface.

[Gewehr98]

Bummer on the battlefield if the poor schmuck sees "Batteries not included"

[onions!]

The exo-armor thing has a science fiction attraction about it, but I'm not convinced of its real world utility.  

Not to put too fine a point on it but,something similar was said about computers a few years back.

[Joe Demko]

Ever watch a kid who's really good at video games play?  They don't have to think about what they are doing.  The "you wear it, not drive it" advantage of powersuits is largely negated by improving the control interface between the pilot and the tankette to where the pilot is able to concentrate on his goals rather than simply controlling the machine.  Powersuits are going to require a high level of AI in order to operate on a "you just wear it" level.  Much lower levels of AI will allow a tankette that the pilot essentially intuitively controls while concentrating on his military goals.

[Joe Demko]

The exo-armor thing has a science fiction attraction about it, but I'm not convinced of its real world utility.  

Not to put too fine a point on it but,something similar was said about computers a few years back.

Computers have, since they were still mechanical, been acknowledged as useful for a great many things.  That's a fact.  I'm still willing to be convinced that the powersuit has some advantage over much less complex (therefore cheaper and most likely more reliable) one-man AFV's.  Other than "it's way kewler" what is that advantage?

[Strings]

Everybody always says "the power supply is the problem". unfortunately, it's only one problem...

 It is ALWAYS easier to make something to punch through the armour, than it is to make better armour. And there are obvious weak points that, the more you harden them, the less mobility you'll have (and you do NOT want to sacrifice that mobility)...

 Think of the groin and armpit: both areas have to keep a wide range of motion, or you risk turning the wearers into the old Star Wars figures. In the case of armour meant for the modern battlefield, that can become a HUGE issue: what happens when a projectile with a high enough velocity penetrates one of those weak points? It bounces around...

 You also then have to make weapons that are "suit-proof". A soldier wearing a strength-enhancing suit of powered armour, upon picking up pretty much any rifle currently available, is gonna crush it in short order...

[Mabs2]

Yea but see, the Gundams were super special hotrod versions of mobile suits.
The mobile suits were MASS PRODUCED.


PS:  Zieg Zeon.

[BrokenPaw]

Ever watch a kid who's really good at video games play?  They don't have to think about what they are doing.  The "you wear it, not drive it" advantage of powersuits is largely negated by improving the control interface between the pilot and the tankette to where the pilot is able to concentrate on his goals rather than simply controlling the machine.  Powersuits are going to require a high level of AI in order to operate on a "you just wear it" level.  Much lower levels of AI will allow a tankette that the pilot essentially intuitively controls while concentrating on his military goals.

True, a non-"just wear it" interface can, over time, become intuitive.  I can control Gordon Freeman with a mouse and keyboard without thinking about it.  But.  Most interfaces other than a suit-type one rely on using manual dexterity to make up for interface limitations.  Most of them don't make much use of other parts of the body (for instance, of all of the controls in a car, how many are controlled by your feet during normal driving ops?  Three, if you drive standard.  Two, if you drive automatic.  One foot is dedicated to (and wasted on) exactly one thing:  Go faster, or don't go faster.  Things like the ability to turn the head and torso as a means of control are often overlooked.  (Yes, I'm aware of gunnery seats that track the gunner's head position for aiming purposes.  That's a good step in the right direction).

As an example, it's possible to control a two-wheeled self-balancing scooter without using the hands, simply by allowing the machine to read subtle cues of stance.  This leaves the hands free to do things like manipulate a tool or weapon.  So there's no reason that vehicles need to be hand-controlled, but it takes a rather a lot more engineering to get an intuitive interface to work, whereas a steering wheel is dirt simple.  But more and more, computrons come cheap.  The stumbling blocks to making intuitive interfaces and wearable powered armor are more in the realm of compact power-plants and size-efficient actuators, which are engineering problems, not computer/AI problems.

I think we'll eventually see powered suits and armor for much of the same reason that the Nintendo Wii got a lot of people into video games that had never been interested before; if the interface is one you don't have to think about, even to the extent of being trained in its use in the first place, then you can get on with the business of getting work (or play) done, not learning how to get it done.

Combat aside, a powered armor suit for SAR would be outstanding; it would protect the operator from hazardous environments, would give greater strength for things like lifting and moving debris, and could get to places that a wheel- or track-based vehicle could not, as well as operating in environments that a helo could not.

-BP

[AJ Dual]

I do agree that these criticisms of powered armor may well be true, and possibly insurmountable, but if you look at what DARPA's funding, they are working on it.

As to "what's the advantage?" I believe that the four percieved advantages of powered armor would be:

- Kinesthetic battle responses. As Joe Demko said, "point and shoot" interfaces are wonderful, but as BrokenPaw points out your own body is the ultimate. If you've tried any video games (such as the MechWarrior) franchise, or tank simulations where you can control camera (head) movment, independantly of your direction of travel, you know how disorienting it can be. However, we do it with our own body every day without a thought.

- A system that can single-handedly bring crew served, or vehicle mounted weaponry to bear in environments that are not 100% vehicle navigable. A "mini AFV" can't jump over a wall, or dive into a ditch either, nor can it climb out of a hole.

- Dexterous manipulation of the environment. Ripping open doors, moving or picking up objects (or people). Vehicles can batter things down, but they can't pull them apart easily unless you get out and throw ropes on things.

- Enhanced survivability from small-arms fire and near misses from larger weapons than infantry with which it (Should, if it's worth it) be able to keep up with, or exceed.

I don't see there'd be any need for "suit proof" weaponry, in a realistic design the weapons would be built into the arms, not held by the manipulators like in Japanime. They would be kept free to deal with obstacles at all times or push the exo-suit back up if it has to "hit the dirt" or is knocked over.

And as far as armor goes, it needs to be determined how much is needed vs. weight and material costs. If an exo-suit is an RPG magnet like a HMMV and light armor are, then I agree, it'll be an expensive failure. However, if the exo-suit can move like infantry, or better than infantry, (admittedly one of many benchmarks needed to prove an exo-suit is worth it...) direct hits won't be an issue, but survivability from fragments from area fire and near-misses will. And I would then expect any worthwhile design of exo-suit to have much, much higher survivability.

The power supply issue is a tough nut to crack, but I think some who propose this roadblock are thinking about powered weaponry AND a powered suit.

Here at least, I'm still talking about conventional weaponry. Aside from sighting systems and electronics, missiles, grenades, and bullets are "self powered". So no multi-megawatt power sources for lasers or rail guns are being asked for. We need power to move, power to communicate, power for environmental control, and power for sensors. The power requirements for gear on that list that infantry already carries are negligible already, so the two main power requirements are movement and environmental control. And that may be within reach of fuel cells, micro-turbines, or even good old IC engines.

"So what if it runs out of fuel?", is the obvious question then.  What is the design's operational envelope? But you could also ask what is the continuous combat endurance of an infantryman NOT in fixed fortifications or positions before he keels over?

Our troops are driven into (and out of) battle, even our tanks are. So it's reasonable powered armor would be as well. As long as it's logistical trail isn't any worse than any other system the military has deemed worthwhile, I don't see it as a huge issue.

Of course, the development of "robust" AI, or even just full-body telepresence might make the need for having a human in the suit unnecessary. And then other form-factors such as "spider/crab", "centaur" etc. for the now "device" (not "suit") are possible.

[Joe Demko]

Let's talk ground pressure (http://en.wikipedia.org/wiki/Ground_pressure)  What do you proponents of powerarmor picture the total assembly weighing?  I can easily see anything but hard, dry ground being the operative eqivalent of quicksand for powerarmor troops.  Sand in the joints...wooded environment...really I don't see the "all terrain" adantages of these things.  Given the mechanical complexity of making the thing walk over a conventional wheeled or treaded AFV, on top of the ground pressure issue, I just don't see these things going too many places a conventional light AFV couldn't.

[Manedwolf]

Full-body telepresence is also an interesting idea. What would you rather send into an insurgent-filled house first, a live soldier, or an anthropmorphized "droid" that can manage stairs, debris and the like, with two hands controlled by a soldier with gloves and movement controllers safe in an armored vehicle outside, handling it as if it were a FPS game in realtime? People are VERY good at that, now...and I can see definite real-world application if the response time is instantaneous and the robot's manipulators are dextrous enough to handle a doorknob. It could have built-in guns, since a drum magazine could be managed, certainly. 

Since they've already got some that can handle telepresence surgery, now, it shouldn't be too much of a stretch.

If you were going to walk into an unknown, hostile building, wouldn't you rather do it with a virtual self first, in the form of a walking, high-dexterity robot with stereo vision, head moment that mirrors yours and weapons?

[Joe Demko]

Actually, remote-presence drones are so vastly superior to powerarmor that I don't see the need for the latter at all.  Far easier to make a drone that is somewhere in the neighborhood of man size and weight than to make something at all  wieldy that has to contain a meat person.  Anything that is "just wear it" controllable is "just wear it remotely" controllable.

[Werewolf]

Anything that is "just wear it" controllable is "just wear it remotely" controllable.

The end result of that is to make war so safe that no one gets killed or hurt which ends up with war becoming pointless or something that's done all the time or...

Well - y'all fill in the blank.

Wasn't it Robert E. Lee that said something to the effect that it is a good thing war is so terrible lest we become too fond of it.

[AJ Dual]

I would say that to keep the ground pressure reasonable, one would try to keep it in the human range of 12psi maximum, (again, the notion of bringing armor and heavy weaponry into the infantry domain...) and assuming a heavily composite structure, keeping the exo-suit under 700 lbs, wearer included, would be a worthy goal. Which would require feet of just under 30 squin each.

Although the ground pressure figure is just a static measure. When stepping it's dynamic. With a peak moment of pressure. Although I'd think that many different kind of animals may do very well in various soft soil environments with possibly higher ground pressure peaks (Hoofed and cloven feet etc...) running, jumping etc.

Collapsing structures you may want to enter for search or CQB purposes would definitely be a concern, but comparing legged locomotion to tracked or wheeled is apples and oranges. You can lift a leg up and reposition it immediately if footing is uncertain, something a wheel or tread can't do.

I do ultimately agree that drones or telepresence may make power armor obsolete by the time it's perfected. Much of the tech would obviously transfer, and confer the weight savings of removing the wearer and all of his support systems from the equation. I think it hinges on how quickly robust AI that's capable evolves. (And if that then drives us to a technological singularity, making warfare moot anyway..) Because the ground combat environment is an exponentially more complex processing task, as compared to what a UCAV must face (even a "crowded" combat sky is a rather empty volume of space...), or say a single-mission/purpose armed swarmbot is expected to do. And in the case of telepresence, robust AI may well still be needed to properly filter the experience for the operator.

If that's not achievable, then a human will need to be in the loop. One possibility I see is having the powered armor wearer being the "king piece" in a chessboard of various drones swarmbots etc.