First Dynamically Balancing Biped Robot

damg writes "Anybots, which is three guys led by Trevor Blackwell, has developed the first robot that walks like we do, by dynamically balancing itself rather than being pre-programmed for walking like Asimo. The video shows the robot walking and being pushed by another 'bully' robot to demonstrate that it can't easily be pushed over."

Oblig

I, for one, welcome our new dynamically balancing robot overlords

He can walk the walk

But can he talk the talk? I didn't think so. Move along, nothing to see here.

Re:He can walk the walk

C3PO: "Oh, R2D2, don't be jealous because *I* can walk. (robosnicker)"

...R2D2 pisses oil all over the floor, watches as C3PO slips and falls on his shiny metal ass; R2D2 emits a sound suspiciously like Nelson from Simpsons "ha ha".

Walks like we do?

Hmm, so, it walks like we do?

Gets out of bed, goes to the fridge, pulls out a soda and pours a coffee, then sits at the computer for 12-20 hours stopping only to walk to the door to receive food deliveries, go to the fridge to get another soda, and to the bathroom to remedy the situation that the previous two types of activities has caused?

Re:

Hmm, so, it walks like we do?

Well, it doesn't walk like I do, but it walks a hell of a lot like my dad does!

Re:

of course, we took major hits at the stockmarket, but hey

      I sold short, you insensitive clod!

It's like falling and missing the ground

This is ways more impressive than you'd guess from the video, which doesn't look much different from all the other walking biped robot video (it's less shiny than most). Since this always pops up [slashdot.org] and always has to be explained [slashdot.org]:

not dynamically balanced:

When the robot (e.g Asimo) moves, it's center of gravity is ALWAYS above the foot it is standing on. As a consequence, the robot could freeze at any moment without falling. Humans can walk that way, but it's slow.

dynamically balanced:

The center of gravity is not above the foot, basically it's falling forward, the motion cannot be stopped without falling. Much faster to move, much harder to calculate. Anybots managed this, which makes their bots a great achievement. We move this way.

Re:It's like falling and missing the ground

This comment is spreading misinformation.

ASIMO _absolutely_ dynamically balances. I have one in my lab at CMU and have worked with walking humanoids [kuffner.org] for years.

First of all, static stability means the center of gravity (CoG) is inside the base of support (the convex hull of the ground contact points).

Dynamic stability is much more difficult to analyze because it involves stability over time. The velocities of any moving parts of an articulated body induce linear and angular momentum that can result in a dynamic stability over time _without_ having any of the intermediate poses being statically stable.

The center of gravity absolutely _does not_ always stay above the support leg for ASIMO. If you try to freeze his pose during the middle of a step it will fall over.

Instead, ASIMO is controlled to keep the Zero Moment Point (ZMP) always inside the convex hull of the contact points. The ZMP (related to the Center of Pressure) is the point on the contact surface where the sum of all torques (moments) is zero. For a given walking trajectory, if the ZMP always stays inside the base if support, then the walk is dynamically stable.

There have been numerous humanoids that use the ZMP formulation to control and maintain dynamic balance while walking (e.g. Honda P2, P3, ASIMO, U. Tokyo H6, H7, AIST HRP2, Waseda Wabian 1, 2, KAIST Hubo, Toyota Partner Robots, Sony QRIO and many more). ALL of these robots are dynamically balancing and are definitely NOT statically stable.

Yes, not the first dynamically stable walker.

Kuffner (above) is right, of course. Dynamically stabilized walking has been around for years. It's not easy to do, but it's been done. Raibert first did it in the 1980s. See his book, "Legged Robots that Balance".

Most of the self-balancing walkers, as Kuffner points out, use a ZMP-based approach. This works for walking, although it's not quite enough for effective running.

Many of the dynamically balanced robots can rebalance after a shove. BDI's Big Dog can. [bdi.com] So can some Japanese hobbyist robots.

If you're not up to date on how far along Japanese hobbyist robotics has progressed, see these videos of this month's humanoid robot soccer match [robots-dreams.com]. These robots are mostly manually controlled, but have computers managing some functions. Many have rate gyros to assist with balance. Gradually, the computers and sensors are taking over more of the control. The hobby robotics manufacturers in Japan now have about 70% of the functionality of Asimo at 2% of the price. There are hobbyist robots with WiFi links and cameras on board. A few more improvements and you'll be able to do all the Asimo stuff with a $1500 robot. But it will only be about 60cm high.

Re:It's like falling and missing the ground

Watch the video again. The cords are hanging loosely and jiggle when it moves. If they were supporting him, they would obviously be taut, like an actor hanging from a wire or a kid on a swing. It makes plenty of sense for the cords to be there to provide power and transmit/receive data. Without the added weight of a battery and computer, they can work on getting the mechanism to work first, then work on getting it to work untethered under heavier weight loads later.

Re:It's like falling and missing the ground

Those cables are used for power and data.

You also may have noticed a mobile frame on casters that surround the robot. Most of the time, I'm sure they connect this robot to a harness so it doesn't fall down. We're not talking about pride here. When your robot falls and breaks something, that shit gets expensive!

Basically, this is like a child riding a bike with training wheels. Just because they are there doesn't mean they're being used all the time at that given moment.

Damn it....

Time to get some Old Glory Insurance, my friends.

Just watched the video and it...

Looks like these robotics enthusiasts had a fair bit of free time on valentines day to put together the video demonstration :)

So what?

I figured that out like, 20 years ago! Kids these days are reaching a new low. They can't even figure out how to walk without some dumb robot to teach them!

Fine, so he wasn't pushed over

Let's see a robot deal with handing over his lunch money or threats of a wedgie. Then we'll have something!

Cool

I immediately noticed a striking resemblance between how this robot was learning to walk and how my grandson, now 10 months old, makes regular attempts to also solve this problem for himself.

Way cool.

This is cool... but the claim of "first" is untrue

This is a cool robot, but the claim of being the first "dynamically balancing robot" is an overstatement. There have been many dynamically balancing robots before, the most famous being Honda's P2 unveiled in 1997. After that, there have been dozens of walking and dynamically balancing humanoids.

What I think the story _should_ point out that is very impressive is:

1) The robot uses pneumatic actuators, which are notoriously difficult to model and control. Almost all of the current dynamically balancing and walking humanoids use electric motors (e.g. ASIMO).

2) Anybots claims to have some "learning" in their controller. Although they don't have any papers about what they are doing, perhaps they are using some clever statistical modeling and feedback to adaptively control and regulate the robot's stability.

BTW, I had a chance to meet Trevor Blackwell a few years ago when he visited my lab. He is definitely a talented engineer with a vision for the future. Several years ago he made Slashdot when he announced his homemade Segway:

http://hardware.slashdot.org/article.pl?sid=03/09/ 28/1756241 [slashdot.org]

From my point of view, any interest by hobbyists and industry in humanoid robotics in North America is great for the field of robotics research.

Dexter and Big Dog

This is obviously a great advancement with enormous potential. But apart from aesthetics I would have thought four legs more practical than two in many circumstances.

DARPA (the US military research folks who helped bring us the Internet) is currently funding Big Dog [wikipedia.org] which I think is has far more potential, because however you calculate it, a quadruped has to have more stability than a biped. Though, in no way do I wish to detract from the achievements of Dexter.

See Big Dog in action here [youtube.com].

Fascinating

Watching the video, I couldn't help but notice that this lacked arms. It strikes me as rather odd to see this, because arms are one of the key features of human balance, but then again they do make for more variables.

Regardless, this is excellent to see.

Re:Fascinating

arms are one of the key features of human balance, but then again they do make for more variables.

Yes and no. You could probably walk or run perfectly well around your house if your arms were completely limp at your sides, or tied tight behind your back, or bound to your sides somehow.

Arms CAN improve balance by shifting your center of gravity quickly, but they're absolutely not involved in locomotion.

Competition

Microsoft is coming out with a robot that throws chairs.
           

Terrible Secret.

Fortunately Shover Robot is there to save him from the terrible secret of space.

Pusher Robot

I like that they seem to have built a whole other robot for the sole purpose of pushing the walker robot.

Though I suppose it's necessary to protect these new inventions from the terrible secret of space.

What, no arms?

Isn't dynamically balancing easier to do with arms? Not to mention, they allow you to move faster and they can catch you if you fall.

I'd bet it would not be too difficult, certainly not easy but compared to their work to this point it'd be trivial, to add some arms and significantly improve the robot's ability to walk... perhaps even jump and/or run... they could even have it catch itself if it falls.

Re:

Smoother? Yes.
Natural? Not in the least.

Asimo doesn't walk like we do. Dexter (this new one) looks like a 10-month-old trying to learn to walk. A robot built to walk in the same vein as Asimo will never be able to walk as easily as a human does. The