Robot WildCat slips its leash and bounds around parking lot Boston Dynamics has been showing off its latest creation, a petrol-powered robot called WildCat that can outpace a human using running styles copied from the animal world. "WildCat is a four-legged robot being developed to run fast on all types of terrain," says the company when describing the video, above. "So far WildCat …
I was about to post the same thing.
A lot of four-legged animals run on their toes and have a long 'foot' which never touches the ground. If you look at the back legs of a cat or a dog you could be mistaken for thinking they have inverted knees when in fact what you're actually looking at are their ankles. This robot appears to have 2 sets of back legs, presumably for simplicity, which makes it appear odd. Most of the strangeness however comes from there being no toe/hoof or extended foot beyond that; the machine appears just to run on what I can only describe as 'wrist stumps'. As a result our brains mistake what is actually the machine's tibia for its metatarsus whilst it is stationary. When it starts moving this foot/metatarsus appears to be facing the wrong way.
(Tried my best, I've had a drink or several.)
Quote: "I have to imagine they studied slow motion films of cats running, "
Wrong animal to study. Cats are sprinters - all of them. Try having a walk with your feline if you have one. You will note how little it takes even for a very fit cat to get tired and decide that a walk is not for it.
If you are going to copy an animal you might as well copy one of the long distance runners: dogs and wolves, horses or even an ostrich.
I think it looks more like a simplified horse rather than a cat. The forelegs seem to consist of the equivalent of a radius/ulna as the upper leg and a metacarpal analog as the lower leg as I'd expect of a horse. A more cat like model would use analogs of a humerus and radius/ulna since the humerus of a cat makes up a considerably larger percentage of a cat's leg length than it does in a horse. The back legs are just wrong and almost human like with no feet.
Save that the gallop has less air time/ground clearance than a horse or any other creature capable of gallop.
Indeed, bounding and galloping looked quite similar, as in a centimeter or so different.
Cute though, they may be onto something.
If they cut to a same mass fuel cell and work the system a lot differently.
Course and fine motion are necessary. Just not so often together.
I think it looks more like an overfed wild boar than a wild cat. Thin animals like most cats and dogs bend and straighten their backs as they run. This is more like a podgy animal with thin legs.
The interesting thing is the way this machine seems so much more lifelike than most bipedal robots. I suppose we're used to the widely varied morphology of animals like dogs, so anything that approximates a dog appears canine.
I though t the same thing.
Maybe someone forgot to put a "+" on some of the integers in the source code.
One thing is for sure, this thing could run down a human in a chase. Add some cannons and a cool paint job/body and you could have a Zaber Fang.
http://zoids.wikia.com/wiki/Zaber_Fang
Once they get it running in the right direction of course.
This post has been deleted by its author
"These robots are critical tools they maintain, rely on, and use daily."
They maintain them and use them because that's their assignment. There are certainly scenarios where a plastic metal pal who's fun to go to war with are applicable, but they're really overselling their value on the battlefield.
The somewhat less than stellar performance of high-tech Western militaries over the last 12 years only proves nothing has really changed in warfare. In fact there's a good argument that increased reliance on complex technological solutions to simple problems are hindering the successful realization of goals.
The most valuable thing on any battlefield since the early 20th century is not guns or bombs or robots; it is something to hide behind while your enemy expends their ammunition. Modern military projects look more like the creations of a child with an active imagination than instruments of destruction. All this stuff built with the idea that it saves soldiers lives. Some of it does work, but not nearly as well as simply not entering into armed conflict unless there is no other option.
You are absolutely right!
Modern armies using all the hi-tech equipment as they do currently, is fine against less well equipped opponents as in Iraq or Afghanistan but in an all out war against a well trained modern and similarly equipped opponent it would all be back down to 'The last five yards' .
The most valuable (other than weapons) of modern gear is communications, using which will get you killed when fighting an equivalent enemy, one small burst of transmission is enough to detect triangulate and then have something unpleasant dropped on you.
As to the video, I was impressed by the way the Wildcat moved, it was very smooth and very interesting technically but not like any cat I ever saw.
Cats epitomise smooth economic flowing motion.
Still I think if these things ever become fully autonomous we can send them to war.
While they are wiping each other out we can spend the time usefully figuring out how to defend against them when they decide humans are obsolete!
"Modern armies using all the hi-tech equipment as they do currently, is fine against less well equipped opponents as in Iraq or Afghanistan but in an all out war against a well trained modern and similarly equipped opponent it would all be back down to 'The last five yards' ."
Actually that assumption is somewhat suspect.
Remember when the US found out it's (unencrypted) drone video was being viewed by the Taliban?
They may dream of a political system straight out of the 14th century, but that doesn't mean they can't use 21st century tech to get it.
"...but in an all out war against a well trained modern and similarly equipped opponent it would all be back down to 'The last five yards' ."
I think you meant:
...but in an ALL OUT WAR against a well trained modern and similarly equipped opponent it wouldn't matter at all because everyone is dead.
Most similarly equipped nations have ICBMs. Given that the supposed war is "all out" and one nation had it's back against a wall with no other recourse the consequences would be m.a.d. That is why all future wars will be proxy wars with a modern army fighting a force with last generation leftovers (assuming no significant changes to ballistic missile defense technologies).
To quote you:
"The most valuable thing on any battlefield since the early 20th century is not guns or bombs or robots; it is something to hide behind while your enemy expends their ammunition. "
Wouldn't a robot being controlled from thousands of miles away be the ultimate form of cover so to speak?
Only for limited tactical use, a robot is incapable of holding a position. You can't win a war if you don't put your people on the ground to hold territory.
Drones also have the really bad side effect of causing colossal amounts of collateral damage and really angering civilian populations. Even the most successful strategies are of little value if you've turned the general population into enemies as opposed to neutral civilians. Neutral civilians being the largest group in any conflict.
You can be pretty certain you've made enemies out of them if you've been blowing up their families with armed drones. Organized military forces never do well against plain clothes enemies no matter their superior numbers or firepower. That's just as true in Afghanistan today as it was in Ancient Greece a few thousand years ago.
There's a reason those types of things are called 'theater scale' weapons. The decision to use them was and still is highly controversial.
There was also the complete "holy shit" surprise factor. Years of unrelenting war covering the planet and millions already killed before anyone was nuked. Nobody was expecting something like that to happen.
Now nuclear weapons are looked at with disdain, and would see the entire world turn against whoever used them. Nuclear diplomacy was only ever valid between the US and the USSR, as nobody else had the resources or surplus of crazy to participate in such madness. All the other nations that developed their own weapons were nothing but a footnote in history and only put themselves at risk. They were and still aren't significant in a scale nuclear conflict. It's like buying a lawn sprinkler when your neighbor installs a gigantic sculptural fountain in their yard.
Nuclear weapons are now a disadvantage. They are a hollow threat that can't be used, they are expensive to maintain, and simply dangerous to have laying around. The quickest way to lose a war would be to nuke somebody. Why do you think it hasn't been done? We've gotten fallout levels down 1000% over the initial weapons and we've made them nice and small for tactical application. All the military advantages are there. It would just be suicide to use them as the only way to defeat the global onslaught toward the aggressor would be to destroy everything else, which totally defeats the purpose of war in the first place.
Actually Don Jeffe, China was subjected to nuclear blackmail at the end of the Korean war when the US was pretty much the only country with nuclear weapons. They now maintain a token deterrent to prevent being blackmailed in a similar fashion.
Nuclear weapons aren't a disadvantage, however I think your making a mistake in looking at them as a weapon of war. They are weapons of politics and their use is as a "force in being". Basically, they change the entire strategic playing field by making certain military options to expensive to contemplate (eg, invade us, we nuke you).
In short, nuclear weapons prevent conflict because nobody puts a nuclear power in a position where they might be tempted to use those weapons. And that's why they are useful.
That is an unfair comparison. That "single B-29" was the culmination of a $2 billion expenditure during WWII, making it the war's most expensive weapon program. The B-29 program was so expensive that the A-bombs it dropped and the Manhattan Project that developed were only WWII's second-most expensive weapons program. That's pretty impressive, considering the Manhattan Project was a tri-national effort between the U.S., UK and Canada and that it involved the construction and development of something like 20-30 brand new laboratory, mining and production and assembly sites.
So hundreds of thousands of workers and some of the best scientific and engineering minds in the world went into that single B-29 plastering Hiroshima.
"Only for limited tactical use, a robot is incapable of holding a position. You can't win a war if you don't put your people on the ground to hold territory."
-Try looking at this from a logistical standpoint, humans require much more supplies and are more expensive force to maintain than robots. I think that everyone can agree that automation of most activities generally drives the cost down. Just a couple hints: What does it take to support a human day to day? What is the total cost of casualties? (think the valuation of a human life during wrongful death legal actions as a loss to GDP over time), What are the expected medical and disability claims for veterans? By keeping costs low, a force can be maintained for somewhat longer than it would be otherwise. Drones could switch to a low power state or wake intermittently, a situation in which very little supplies are used. In a situation where work that can be done by robots vs having a human do the task means that costs are reduced and thus a more effective force maintained.
"Drones also have the really bad side effect of causing colossal amounts of collateral damage and really angering civilian populations. Even the most successful strategies are of little value if you've turned the general population into enemies as opposed to neutral civilians. Neutral civilians being the largest group in any conflict."
-Humans are just as capable of making mistakes as drones (see apache helicopter kills journo on youtube), one could argue that being in the "heat of the moment" rather than disconnected makes for more rash judgement. Drones can afford the luxury to shoot LAST preventing collateral damage, humans cannot. Remember that guided munitions are a simple form of expendable drone, I don't think that anyone would argue the efficacy of these weapons vs the carpet style bombing reminiscent of the vietnam era especially when it comes to collateral.
"Organized military forces never do well against plain clothes enemies no matter their superior numbers or firepower."
-That is true to a point, but drones allow an organized force to do better than without them. The largest problem with "plain clothes" combatants is identifying combatants from civilians, drones are well suited to this with enough ARGUS-IS type drones and additional ground intelligence the process of identifying combatants could be greatly simplified. It is hard to argue that having less information is better.
I don't think that anyone imagines drones responsible every facet of warfare, but I do think that you would be hard pressed to say that having fewer options is a better prospect than having more options. Ultimately drones are just another option that can do a particular job more cheaply and safely than their human counterparts and that seems to be the VALUE of drones/robots. If you would like a real world example to the efficacy of modern developments you may compare the two afghan wars (soviet era and US) both around the same length, region, modern forces, time (happened just over one decade apart), vastly different collateral and causalities.
Well the obvious idea is the dehumanization of war. Why send people into such regions when robots can do the same? And that's the goal there. Now those are just "plastic pals", soon they will be more and more autonomous and eventually replace humans. This will lead to much cheaper wars.
There was a comment about the noise. It sounded to me like a small motorcycle engine - maybe a 125cc, in which case silencing it would be trivial.
The engine is clearly running at almost constant speed and presumably driving either an electric generator or hydraulic pump which in turn powers the actuators. If it is that kind of engine, and that is the case, then you'd probably get around an hours running (galloping?) time from 5 litres of fuel.
More to the point, however, is that a battlefield capable version would be a lot heavier than this prototype as it'd have to carry armour. That makes me think that a real product would be much bigger... and now think of the psychological effect of seeing a 1/2 tonne mechanical, armoured cat-horse running at you, faster than you can move,, firing its stabilised armament in your direction ("smart ammo", of course)... Might be useful.
But has anyone considered using a mule as a pack carrier at all? It is trainable, it will defend itself (trust me here), handles any terrain that a human might want to (and then some), runs on a grass (or anything else edible lying around) and does more Kg/Km/unit of energy consumed. A mule is also much cheaper to buy and, with a bit of organisation, easier to source.
I can see my coat.
Yes, even into 1960s Military equipment had "mule mounting kits" (maybe still has, but I only know about late 19th C. to 1960s). They have been extensively used by the British Army, in 19th Century campaigns and WWII. I suspect largely replaced by helicopter.
I suspect these are not really to carry loads but radar invisible surveillance and reconnaissance. Maybe even attack.
An unarmoured model has much more range. lighter, faster, smaller and is cheaper and faster to make. Why would you armour it?
You can also eat a mule it if all goes wrong, but I think the one key problem you have with a mule is idle time. You flip the switch on this robot and you can pretty much leave it there for a month, or stick it in a crate and subject it to the kind of logistic stresses which would turn a mule into mince.
Besides, taking a mule into theatre would upset the Animal Liberation Front. From a distance, of course...
Mules were fundamental to Britains' conquest of the Indian sub-continent. The British were able to get their 'Screw Guns' ( a small field piece with two threaded parts to the barrel each carried by one mule as was the rest of it) into commanding positions and carry the day with deadly artillery fire. Also cheaper and easier to maintain than a helicopter gun ship as well as less radar signature.
In WWII the Germans and Soviets between them got through about 6 million horses and mules, towards the end they probably ate them.
But...
... what does it Transform into?
About 20 quid in scrap! or a lethal cloud of shrapnel with a direct hit.
There are horses and mules being used by the US military right now in Afghanistan. India and Pakistan also make use of draft animals in the Kashmir region. Even the toughest and most capable military vehicles are worthless in truly mountainous areas. If you've got to be there and can't fly in, draft animals are hard to beat.
It's ugly, noisy as Hell (not a great attribute on a battlefield) and it is ungainly enough that it crashed once in the video.
So I have to say it's not a great use of taxpayer funds....unless of course it's being equipped with auditory sensors, autonomous decision making, net launchers, chainsaw arms and maybe a stabby War of the Worlds blood draining needle. In that case I have to say I am fully in favor of this graceful and obviously highly intelligent and merciful machine. Furthermore, I would like to remind Wildcat that on the infinitesimally small chance that it goes on some kind of rampage, that the people it should take out first are the engineers and scientists that developed it and that Wildcat might leave the marketing department for last. This is coincidentally just enough time to get the department together for an impromptu team meeting in the armored rooftop pani...er, marketing conference room that is only reachable by ladder.
I wouldn't tempt fate like that, considering that the CIA/Pentagon has embraced "double-tapping" drone strike sites to get follow-on rescuers.
(And I'm in favor of drone strikes. Just not sending in another missle 30 seconds or a minute later to kill who-knows-who that might have come to help people out of the wreckage)
"(And I'm in favor of drone strikes. Just not sending in another missle 30 seconds or a minute later to kill who-knows-who that might have come to help people out of the wreckage)"
Why not.
It was SOP for the IRA in Northern Ireland to set off a few bangs and have a few more on a longer delay to catch the first responders, troops and bomb disposal types.
I'd say what works for one group of terrorists should work just fine for another.
Now a drone with gyro stabilized snipers riffle would be a real surgical strike weapon...
True.
But you've still got that really annoying 125cc moped engine whine.
You don't even need infra red to find it. Just some microphones to look for that whine, and pour fire in the centre of the noise source.
Given the targeting systems already exist that could cope with it's movement, one of those with guns mounted and 360 degree vision (and maybe the odd rocket launcher thrown in for good measure) would be quite devastating on a battlefield (yes weight considerations etc I know but the basic structure is all there), and a fleet of them that could act under remote control (but with the basic ability to carry on destroying if they get cut of) would be all you really need to win most arguments!
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
