New BigDog Robot Video

John860 writes "The US company Boston Dynamics has released an amazing new video of its quadruped robot BigDog. The highlight of the video (at 1:24) shows how the robot starts slipping on ice, almost falls several times, but finally regains its balance and continues walking. The video also shows the robot's ability to cope with different types of terrains, climb and descend steep slopes, and jump. Two years ago, the older version of BigDog was already able to climb slopes, keep its balance after a strong kick, and walk on rough terrain like stones, mud, and snow. The new version weighs 235 lbs and can carry a payload of up to 340 lbs, a factor of 4 better than its predecessor."

Comments

Re:Somewhere deep in the caves of Tora Bora

[Hal_Porter]

Did you ever see a BBC documentary on putting explosives inside animals? As you can probably expect, spooks experimented on it during the cold war. Partly it was because the Russians trained dogs to sleep under warm tanks, loaded them up with exposives and sent them toward the German lines in World War II. The US/UK were quite reasonably concerned that a "explosive dog gap" might open with the Communists, so they poured money into research.

My favourite part was where some scientist enthused that "you can fit quite an arsenal inside a 500lb boar". What an awesome job. I reckon animals are quite redundantly engineered, so you could take quite a lot of guts out of them and still have them able to stagger a few miles to enemy lines. Boars are unclean animals for Muslims, so presumably be drenched with bits of exploding boar milliseconds before you die would stop you getting into paradise if you believe in that sort of thing.

That would be culturally insensitive of course, so we shouldn't do that. But you could turn a herd of goats into a living cluster bomb. Can I has DARPA funding now?

Re:Simply Amazing.

[Anonymous Coward]

If you "don't see any actual use in war, besides transporting things", you're really not trying.

Add a turret, a video camera, and a remote control -- presto, a soldier that can march 24/7 across the desert, across the ice, through tear gas clouds, through radioactive fallout, and arrive somewhere all fresh and ready to shoot people, or drop bombs.

They're not going to "make toy versions", at least not any time soon. Why try to make a $100-1000 toy, and compete on the free market, when you can keep everything secret and sell them to the military for orders of magnitude more?

I'm an American, and these things scare me. Robert E. Lee once said "It is well that war is so terrible, or we should get too fond of it". Our government is making it significantly less terrible (for its own soldiers) all the time, and they also seem to be growing rather fond of it. When you can run a robotic war (in the air and on the ground) by remote control, what's to stop you from attacking everybody you don't like?

I predict we'll have robot infantry on the ground inside of 5 years, and within 2 years of that, they'll be back here patrolling American soil. And no, it's not a partisan issue, either: even Obama, the democratic frontrunner, wants to *increase* military spending, even though America's military budget is already larger than the military budgets of every other country in the world, combined.

Re:Simply Amazing.

[Rainer]

...the dog could be destroyed with something like a bomb or a mine ...

Soldiers can be killed/wounded with the same weapons.
If a robot is cheaper than a dead/wounded soldier the robot might be a better option.

Also consider that robots need no training and (almost?) no supplies when they are in storage.

Re:Cool, yes. Useful?

[stiller]

This is a pretty cool tech demo, but at the moment, its battlefield utility is zero. That two-stroke engine buzz is going to alert every bad guy for miles around. You assume that they'll use it for stealth operations. Not everything on the battlefield needs to be stealth. A tank is pretty noisy, still, it has it's place. For example, in a forrest situation, you might be able to hear it, but you won't see it until it's in a line of sight. And then it's a matter of your reaction speed versus that of a robot. Also, you could simply flood a battlefield with these things - think thousands - and give them all an explosive payload. You just got yourself a thousand kamikaze dogs (or more accurately, locomotive claymores).

No tactical need for anti-tank or self-detonating

[patio11]

There is no reason to use a robot to deliver an anti-tank round when a) the enemy doesn't use tanks and b) if he did, we have 46,000 cheaper, more reliable, and less risky ways of killing the tanks. Similarly, explosive robots have all the ROI of "firing a $2 million missile at a $10 empty tent and hitting a camel in the butt"*, which we have been trying to get away from.

* Best Dubya line ever. http://www.snopes.com/rumors/bush.asp

Re:Cool, yes. Useful?

[hcdejong]

Me, I'd go with a real live mule instead for all applications you'd use this in. Same payload capacity, not much bigger, totally silent, self-refuelling, costs $hundreds rather than $hojillions. Self-refueling? That rather depends on the terrain you're on. Totally silent? Until it brays (or whatever mules do) at exactly the wrong moment and ruins your ambush.
Livestock needs to be taken care of every day, is much more maintenance-intensive than anything mechanical. It also can't be stowed in a container for easy long-range transport.

Career regrets

[ShannaraFan]

That is just so friggin cool. Seeing stuff like this makes me regret spending 20 years sitting in various cubicles twiddling database bits...

Re:Simply Amazing.

[sloth jr]

If you "don't see any actual use in war, besides transporting things", you're really not trying.

Add a turret, a video camera, and a remote control -- presto, a soldier that can march 24/7 across the desert, across the ice, through tear gas clouds, through radioactive fallout, and arrive somewhere all fresh and ready to shoot people, or drop bombs.

And this could be likely achieved with other conventional robotic conveyance mechanisms. If you just need to deliver a mobile land-mine, adaptation of simple RC cars could probably serve. As for dropping bombs and shooting people - there are plenty of airborne weapons that would be difficult to surpass in terms of "efficiency". Cheaper and simpler will win.

About the only military use I can see for this might be urban alley crawls, where terrain could be difficult, cramped, and dangerous, and possibly IED detection/detonation. I agree with parent about this being mostly a pack mule.

Re:Simply Amazing.

[Rastan_B2]

I hate to say this - but thats only modern (developed country) warfare where it would be cheaper, and need no training. I'm sure through many a war in our inglorious past we have sent many, many men to die very cheaply with very little training...

Re:Simply Amazing.

[Kjella]

If a robot is cheaper than a dead/wounded soldier the robot might be a better option. I haven't checked the costs of raising a kid and training him until he can die as a soldier, but given the other expenses in war I doubt that's really what matters. It's the political cost that matters, and it's far far higher. If no US soldiers at all had died or been wounded in Iraq, I doubt there'd be nearly as much fuzz about the war even if it cost twice as much.

Re:Simply Amazing.

[ucblockhead]

Given that soldier's packs have been steadily growing and are now on the order of 80-100 lbs, just being able to transport things is damn useful. A few of these with a squad would mean that the squad could move faster and bring more weapons, more food and more high-tech gear. Having infantry that moves faster and can last longer without having to be supplied would be extremely valuable.

Reaction time

[bar-agent]

I noticed that the reaction time while it was recovering on the ice didn't seem much different from animals that I've seen slip. But you are right that the robot's precision is a lot better; its legs aren't getting in each other's way.

I wonder why the reaction time is about the same. Does the dynamics planner take that long to figure out what to do? Are the actuators slow enough so that it can't recover in a blur of leg motion? Or is that just the minimum amount of time stabilization can physically take?