Slashdot Mirror
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."
The US company Boston Dynamics has released an amazing new video of its quadruped robot
The walking motion is much like a goat. A goat, see?
The theory of relativity doesn't work right in Arkansas.
Hafez, you get the shipment of Korans from Amazon yet?
Yeah, but it was delivered by this weird mechanical goat thing that buzzed like a swarm of bees in a poppy field.
Hmm. I believe my RealGoat delivery has arrived! Allahu Ackbar!
Is anyone else creeped out by how natural the movements of this robot are? Maybe it's the lack of a head and the ominous buzz-of-death, I don't know. As I recall, there's some theoretical curve for robots where the human acceptance of a robot dramatically drops at a sweet spot as reality is approached and doesn't rise until reality is achieved. This robot definitely falls in that zone for me.
I swear to God...I swear to God! That is NOT how you treat your human!
It reacting to a kick was so lifelike I wanted to call Peta. I frankly don't see the actual use in war, besides transporting things, I can't wait till they make toy versions.
I highly doubt that it's very real.. You see, I would kick our dog right now like that.. there!.. his true reaction wou3eim em,,yuktie2;'36+ .0
One giant leap for Imperial walkers...
Robot locomotion of that quality is probably one of the most difficult problems to solve - the robustness of that thing was quite impressive - it survived rubble, snow, ice, and a solid kick that sent it tumbling. I'd really like to know how they did it, if they just managed to perfect current techniques with enough DARPA money or came up with something new - I would imagine it required some very accurate sensors and actuators, and a super-high-precision inverse-kinematics solver. If they can couple that together with a super-accurate local navigation system - which I imagine would be the easy half in comparison - then they've got a huge platform to launch consumer-grade robots if they get to a low enough price (and they do something about the noise). Maybe I will have a robot butler in my lifetime, but it looks like the military gets their mules first.
Maybe the combat versions come with a Lunge and Bite Service Pack that corrects the "Unexpected Response to Kick" bug.
The civillian versions will be all plush and lovable (though huge) until some glitch reenables the combat subroutine. Lone cop and beautiful female computer scientist will then need to fight their way to the Mans' Best Friend central computer to press the reset button. One of the dogs will stay loyal and help them, the rest (with glowing red eyes, to tell the slower audience members that they are Evil) will terrorise the population.
Joe Dante will direct "Mans' Best Friend" (working title "Pastiche 3") of course, from a novel by Steven King. The cast will all be scientologists and there will be a few references to engrams and so on in the script, or maybe just adlibbed in. The movie will start with Eisenhower's speech about the Military Industrial complex and then cut to something ironic, like a weapons factory stripping the weapons off a giant robot dog endoskeleton, wrapping it in plush fur and loading it into a box labelled "Mans' Best Friend".
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
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.
Since it needs to be able to exert pretty big forces very quickly, I doubt they're going to lower the power requirements, so I highly doubt they're going to be able to use a quieter power source like batteries or fuel cells. Nothing beats the power-to-weight ratio of internal combusion.
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.
That is just so friggin cool. Seeing stuff like this makes me regret spending 20 years sitting in various cubicles twiddling database bits...
This is very nice work. It's good to see Raibert doing robotic locomotion again, and finally, with a big enough budget.
Back in the 1980s and early 1990s, Raibert headed the MIT Leg Lab, which produced the first legged robots with real balance control. Raibert started with one-legged hopping machines, to force the balance issue. His big insight was that balance is more important than gait. In 1992, he left MIT and did a startup, Boston Dynamics, and went off into simulation. Most of the simulations weren't dynamic, just kinematic. Now he's back to robotics, and dynamics, again.
I've worked on control of robot running on rough terrain. So I understand the problems. Watching the Big Dog video, I have a reasonably good idea of how it works. This is quite impressive. DARPA got its $40 million worth.
First, it has slip control, like automotive ABS, for its feet. The first insight on the hard cases for locomotion is that balance is more important than gait. The second is that slip control is more important than balance. The key to slip control is keeping the transverse forces at foot-ground contact below the point where the feet break loose. ("Inside the static friction cone", for those familiar with the terminology.) Watch it move on ice. The feet do not slip at all unless there's real trouble, as when someone kicks the thing. The transverse forces are being held below the break-loose point. Given the load on the foot, the actuator forces (hydraulic cylinders on Big Dog) must be coordinated to keep the transverse force below the ground coefficient of friction times the longitudinal load. Finding the ground coefficient of friction can be either trial and error (if it slips, reduce the value) or they may have actual slip sensing in the foot, like humans and animals. Humans, incidentally, tend to maintain a contact force about 20% above the break-loose point, as a safety margin.
Big Dog's reaction to a slip is to immediately raise the foot and go for a new foot placement. That's an emergency behavior, though; it's the prevention of slip that makes it work. Watch the robot's reaction when it slips on ice, and, once you know what to look for, you'll see how it does it. The first priority is to recover traction. As soon as a foot slips, it's lifted and placed in a new position. The second priority is to recover balance. As the robot starts to roll to the right, it executes a violent twist to the right and throws out the right front foot. It needs a foot position within the traction limits to provide the roll moment needed to recover balance, and it has a good enough planner to find one. Look at that sequence and ask yourself first "where does the foot need to be to get traction", then "where does the foot need to be to recover balance". Then you'll understand how it works.
Big Dog has, finally, true gaitless locomotion. Decades of locomotion research have focused on gait, foot sequence, "central patten generators", and similar mechanisms that deal with the easy cases. Wrong answer. The right answer is to think of legs as assets that can be deployed to maintain slip and stability criteria. It's very clear that Big Dog does this; it can use its feet (and knees!) as necessary. It's not constrained to a gait pattern at all.
There's a true dynamics predictor and planner in there. This is not just a reactive robot, like Brooks' little machines. Nor is it a straightforward ZMP ("zero moment point") stabilization system, like Asimo. (Think of ZMP as a generalization of center of gravity to include momentum.) There's a planner with a horizon of (I think) about two foot placements ahead, and it has "what if" internal simulation capability. That's why this robot moves so well. It can predict, at least approximately, what's going to happen for its next move, and plans on that basis. That's why its movement are so smooth. Without that, you'