Honda Creates Walking Robot 130
ilyah writes: "The ever-crafty Honda engineers have apparently created a robot that can accurately walk on two legs (i.e. just like humans). Take a look at The Honda Japan site -- a Japanese site that has some pictures and videos worth looking at (hint: videos are under linked marked "04"). It's pretty awesome watching that thing walk around." If I remember correctly, they've been working on this for quite some time - like 15 - 20 years, but this is apparently one of the breakthrough steps, pun intended.
Hmmm (Score:1)
Re:Hardware vs. clever algorithms vs. refined hack (Score:2)
Re:Sony too. (Score:1)
Here's a link [tbs.co.jp] to a movie showing the little guy do his stuff.
The Sony Robot that is.
Enjoy!
Info and Movies in english (Score:3)
This was posted a while ago, but with a different site. Try this site: http://www.honda-p3.com/ [honda-p3.com] for english text and movies and such.
Successor of the P3 (Score:1)
From what I've heard, Honda is interested in developing a market for robots to take care of the elderly. Making the robot lighter is a good step in that direction. (You don't want your robot to stumble and crush you.) This new robot is walking a lot more fluidly. I think the weight reduction has a lot to do with that. Sony, on the other, is interested in bipeds simply for the entertainment value. The Sony videos are great.
(Here are the emails)
***
Yesterday in Japan, there was a sensational press release of the Honda's new humanoid called "ASIMO" which stands for "Adaptive Step in Innovative Mobility." This robot is a new version of Honda's humanoid robot P3. According to the newspaper article, its height is 120cm and weight is 43kg. Specifically, the weight is significantly reduced compared to the previous one. It is equipped with tendon-driven five-fingerd hands. In addition to straight walk, it is even capable of walk on a figure eight by changing direction without stopping.
***
Sony also developed a small biped robot called "SDR-3X." The height is 50cm and its weight is only 5kg, and this robot has 24 joints. This robot is capable of walking at the speed of 0.9km/h (15m/min), turning aroud, standing up, balancing on one leg, dancing, and kicking a ball. Basically, this humanoid robot uses the same computer architecture as used in AIBO. However, they developed new actuators and motion control architecture for biped walking.
It seems that they are thinking of selling the robot in the entertainment market in the near future at the price of a compact car.
Wasn't this featured in Wired Magazine... (Score:1)
http://www.wired.com/ wir ed/archive/6.11/wired25.html?pg=12 [wired.com]
Rampy
Honda's long-running research (Score:4)
$ man reality
Old news (Score:3)
The more interesting thing that I got from the special was the robot that *learned* to walk like an infant does. At power up, it is wobly and nervous. As the instructor helps it maintain its balance and catch it falls, the robot's AI (neural networked?) learns from its mistakes and improves. After a few hours, the bi-pedal contraption is able to walk on its own. That is, until it is turned off and its memory gone.
Re:Priorities (Score:1)
Hook, line and sinker.
awesome! (Score:1)
VTEC? (Score:1)
Sorry.
- Mike Hughes
The next step (Score:2)
question (Score:1)
FluX
After 16 years, MTV has finally completed its deevolution into the shiny things network
Old, old news... (Score:1)
Simple Pleasures. (Score:1)
Java Gundam at kazzuya.com/test/gundam (Score:1)
http://kazzuya.com/test/gundam [kazzuya.com]
Sure beats the MIT trash I've seen (Score:2)
Mastication (Score:1)
Re:Honda (Score:1)
Because they can? Every development starts with a small step. The world's fastest and largest planes developed from the Wright Flyer. The world's largest ships developed from a dug-out canoe or a couple of logs lashed together.
Who knows, humanoid bipedal robots developed from these may be able to enter places too dangerous for people and save lives. - firefighting, high radiation levels, etc.
I find it fascinating. So long as they don't fit one with a personality prototype, that is...!
Hacker: A criminal who breaks into computer systems
Re:Mastication (Score:1)
Re:Priorities (Score:2)
I completely disagree. Diverse research makes for better answers. I agree that the problems of poverty, wildlife, etc are more important than the puzzle of a walking robot. However you never know when research in another field will turn out to be very usefull in another field. Science is full of times where the best use for something isn't what it was designed for. Diverse research leads to a larger body of science for the people in the "important" fields to draw on. So this robot research may seem frivouls(sp) to you but the resulting body of knowledge may have something in it to help with micro surgery techniques.
For Good, Not Evil (Score:1)
old news (Score:1)
Repost.... (Score:1)
Re:Hardware vs. clever algorithms vs. refined hack (Score:5)
The Honda robot [honda.com] qualifies as an "impressive start"... its 2.0km/h speed of the Honda robot is not impressive for a lightweight tri or quad walker, but it is for a bipedal robot of human size.
The fact that it can climb stairs is especially cool, given the extensive rebalancing done every moment in a step, and the feedback sensors needed to read these motions properly.
Certainly there are lifting robots which can hoist many tons, so the 5.0kg/hand weight limit seems skimpy - but not when considering that this machine can allegedly walk and perform complex arm manipulations while holding this weight. Sadly, its continuous runtime before recharge is only 30 minutes, but I suspect later versions will take advantage of increasing innovation in charge/weight ratios in batteries, and perhaps solar panels for space use (an obvious application of these robots would be EVAs for the ISS [nasa.gov] or other craft).
Cool, in terms of integration with other systems, is the use of wireless ethernet as the comm standard, rather than some proprietary system. This means this robot could be controlled by base-station systems of arbitrary complexity - including a Beowulf [beowulf.org] cluster running a complex AI system like Webmind [webmind.com]. This means that while technology is not quite there yet to put any advanced computational intelligence inside a biped robot, it can be controlled by advanced systems running at fixed-position stations through LAN technology - a good compromise in terms of merging the state-of-the-art in Robotics with AI to try to build towards a better convergence.
Regarding the robustness, it appears that the 25 minutes of runtime is the primary limitation in terms of continuous operation - there is no data I was able to find on failure rates or the fault tolerance of the sensors or computational systems on-board.
As for hacks vs. new general purpose algorithsm... They obviously do not reveal tremendous amounts of details, but suffice it to say that the engineering done to build 3 successive models of bipedal robots that can walk and climb (stairs, hills) represents fundamental work in robot dynamics engineering which, while parameter tweaked for this robot's operations, is certainly applicable (with some tweaks or modifications, as with all engineering techniques) to other bipedal robot applications.
The wireless lan comm technology, improved user interface (over the previous version), and sensor systems are all also certainly reusable in similar robots (indeed, likely also in multi-legged robots).
However, as it is a commercial product AND I do not read Japanese, I was not able to find any papers on specific algorithms to give a more detailed analysis...
Here are some useful resources I did find:
The official site in English [honda.com]
An article about the robot's walking functions [panmanee.com]
Images of the robot at UIUC [uiuc.edu]
Biped Robots in General [caltech.edu]
Robodex Robotics Conference [robodex.org]
Re:Old news (Score:1)
This would not have been a problem if they'd built a bicycling robot...
This story was on the local TV news; IMHO, the robot is a bit too human-like - put a small person in a space suit, teach them to "wiggle" in a cute, AIBO-like manner, and could anyone tell the difference? (I'm sure there was/is/will be a law against this - Bladerunner, anyone?)
Oh my god this is old news! (Score:1)
Re: VTEC? (Score:1)
Of course. Didn't you see the big spoiler and the fat tailpipe on it? :)
--
Turn on, log in, burn out...
poop (Score:1)
Re:Hardware vs. clever algorithms vs. refined hack (Score:1)
http://www.honda-p3.com/english/html/frameset2.
that points out that specific fact, in eengrish even.
The Honda bipedal robots are the most sophisticated on the planet. You might not look so foolish dogging on them when you come up with something better.
In the old days... (Score:1)
I remember about 4 years or so ago, seeing QT movies on the HONDA site of this model's grandfather which was just the waist and two legs moving around.....they hadn't made or put the torso on yet....it was awesome, watching two robot legs (attached to the waist of course) walking up and down stairs and kicking a ball and even running!!!!
GO HONDA!!!! my dreams are in your hands!
Jak Din
Re:Asimo? (Score:2)
Maybe not. "Ashi mo" (which can also be romanized Asimo) means "Legs too!" in Japanese.
Cheers,
-j.
Re:Asimo named after Asimov ? (Score:1)
"Advanced Step in Innovative Mobility"
Don't believe me? Check out this image:
http://www.honda.co.jp/ASIMO
Jak Din
RoboCup Challenge!!! (Score:1)
1. http://www.honda.co.jp/realse rve r/video/robot64.rm [honda.co.jp]
2. http://www.honda.co.jp/re als erver/video/robocupdown.rm [honda.co.jp]
later,
Jak Din
Re:poop (Score:1)
So would you if you had that power-pack-backpack strapped to your back!
:)
Jak Din
Re:Asimo named after Asimov ? (Score:1)
Damn!!! I should be hung for that...the proper spelling is: Isaac Asimov
sorry Mr. Asimov
:(
Jak Din
Walk this way, talk this way (Score:1)
Re:Priorities (Score:3)
But, on a more serious note: What happens if Honda gets to the point where they produce a robot capable of digging wells and planting crops in underdeveloped, potentially hazardous countries? What happens if Honda manages to develop a robot that can act as a lab assistant, routinely, tirelessly and accurately performing a wide range of menial yet necessary tasks in a lab environment? Sure, it's too advanced to be able to accurately predict something like this coming along today, but the very essence of science is the pursuit of being able to do and understand things we currently cannot do or understand. Today, this thing is just a prototype that can saunter across a studio floor. Another fifty years, it may be able to run for weeks on end with minimal human intervention, able to do a huge array of tasks that would require fairly heavy human intervention today.
Just as the pioneers of computing could have no way of accurately predicting the impact computers would have on the world fifty years later, the Honda engineers can have no way of knowing what will become of their robot in another 50 years. Rest assured, though, that they, like the computing pioneers, are serious scientists and serious dreamers...
$ man reality
Re:Sony too. (Score:1)
whats the point? (Score:1)
Book recommendation (Score:2)
Salon.com published a review [salon.com] recently.
Lars
__
Re:whats the point? (Score:1)
Instead of a robot, turn it into an exoskeleton for the disabled. Kind of like that smart wheelchair described in Wired. Of course, it should be able to fire small missiles and jump over at least 10 feet of razor-wired fencing.
it only looks smooth. (Score:2)
By the way, I would like to see those new movies but I can see only mpegs on my linux machine. Does anybody know how I can see quicktime movies on linux?
Terminator (Score:1)
sex robot (Score:1)
Intermediate forms (Score:1)
Re:And what is the robbot saying ? (Score:1)
Re:Hardware vs. clever algorithms vs. refined hack (Score:1)
he, funny, I don't feel foolish.
Suppose it's possible that I know something you don't know?
nah, never.
____
Biohazard could still take it. (Score:2)
Re:Asimo? (Score:1)
Re:Asimo named after Asimov ? (Score:1)
Names are cooler when they have multiple meanings:
Re:Why I think it's a hoax (Score:1)
Re:Why is walking so hard? (Score:1)
The human foot does an insane number of adjustments every second to keep you standing on it. You're right that emulating the inner ear's balance sensor would solve the problem. But that means you have to understand the inner ear (and what the brain does with that data) first, and we're not there yet.
Re:Why is walking so hard? (Score:1)
But that means you have to understand the inner ear (and what the brain does with that data) first, and we're not there yet.
Hmmm. You know, standing on one foot here and thinking about it, it makes me wonder if the inner ear is more of an accelerometer rather than a gravity sensor (which is what I've always assumed). It's a *much* more complicated problem to take into account head alignment and figure if you're tipping than to just figure out which way your head is moving back and forth.
Actually, I think it might be both. If I get grossly out of alignment and am about to fall, it feels "different" than when I'm making only fine balance corrections. I bet gravity is used to get us close to upright, and then acceleration is used to keep us there.
I bet it will take a combination of both approaches. There are laser-based acceleratomers that are extremely accurate. I wonder if anyone has tried using both?
--
Sony's 2 legged robot (Score:1)
The page is in Japanese, but you can still see some cool pictures. Check it out! [aibotown.com]
Ah.... (Score:1)
(Joke blatantly 'borrowed' from last night's Tonight Show.)
---
One cool thing... (Score:2)
Re:Priorities (Score:1)
No it's not. These extremely rare cases you speak of just make better headlines than small and incremental milestones achieved in more important research.
Diverse research leads to a larger body of science for the people in the "important" fields to draw on.
If scientists can even make the association. With so much superflous, corperate-driven research going on around the world, finding an answer to an extremely complex problem (in a radically unrelated study no less) is like finding a needle in a haystack.
the resulting body of knowledge may have something in it to help with micro surgery techniques.
I was thinking prosthetic limb enhancements was a more logical correlation, but i'll buy that.
heh (Score:1)
Stileproject.com reported this already. Of course, Jay Stile added in the comment of "Shit eating capabilities to be added in later."(big inside joke)
ok... so it walks... (Score:1)
Re:Sony too. (Score:1)
Asimo's true identity (Score:3)
Re:whats the point? (Score:1)
Old news... (Score:1)
1st Precept of All Applied Technology (Score:2)
I can imagine the fake Visa adds already:
Tube of hydraulic fluid: $15
Honda robot: $9000
Your name becomes a sexual verb: Priceless
snore (Score:1)
Sl ashdot is becoming worse than the BBC for repeats
Re:whats the point? (Score:1)
The world is built for humans. Cars, buildings, everything. It's all designed to fit our bodies and to fit the way we do things.
By developing really smart robots that have the same shape and movement as us, we can then make all of our other machines cheap and stupid. We will then have one really smart and expensive robot that can independently operate any number of cheap, stupid machines. An example is a house robot: One smart humanoid house robot could vacuum, dust, do the laundry, dishes, answer the door, etc... using the same tools that already exist today and are designed for our use. Robots are plug-n-play. Also, psychologically, people can more easily anthropomorphize a humanoid robot and will be more easily accepted by society at large.
Steven
Re:it only looks smooth. (Score:1)
This would explain why it looks like a total hoax, with a crouching guy in there. (I still want to see proof that it's not.)
But if it's just 'playback' then it's not even AI-- it can't possibly adapt to changed terrain... right?
Re:Asimo named after Asimov ? (Score:1)
Re:Very Old News... (Score:1)
Re:Isn't this old news? (Score:1)
Re:Robot (Score:1)
Re:Robot (Score:1)
Re:Why is walking so hard? (Score:1)
I agree with you that there is more to this that meets the "eye", especially when it comes to anticipating terrain (pun intended). Trying to walk with your eyes closed tells you how much of a visual component there is that you don't realize.
See this message [slashdot.org] for some thoughts I just had about the balance sensor problem...
--
Re:Hardware vs. clever algorithms vs. refined hack (Score:1)
Or don't...and know that all you're doing is wanking on Slashdot...
Re:Asimo named after Asimov ? (Score:1)
I don't really know about ASIMO meaning "Legs Too".....in Japanese, the word for 'leg' is ASHI (but I know that Japanese always make the mistake when spelling in Roman Letters), so you might be right.....
:)
Jak Din
Re:Intermediate forms (Score:1)
hmm. could be, but I'm going to the Robot Convention in YOKOHAMA this weekend, where they are going to/are showing off all the various robots from Sony, Honda, etc.....so when I'll maybe post something about.
Jak Din
A walking robote is fine... (Score:3)
But can it flail its arms around and say "Danger! Danger Will Robinson!" That is the true mark of a great robot.
Re:Asimo named after Asimov ? (Score:1)
Re:Old news (Score:1)
Re:Honda (Score:1)
The idea of a humanoid robot is one that has gripped our culture with fascination for nearly a century (Metropolis, Star Wars, etc.) The reason these robots don't exist is because the enormous amount of development needed to create even a crappy one served as a barrier to entry.
Apparently Honda sees this as a growth industry in the future, and wants first mover advantage into this field. Who wouldn't want a monopoly on artificial slave labor? Or being the sole supplier of mecha-infantry to the risk averse US Army?
More details in English here (Score:2)
I'll buy one... (Score:1)
Re:Honda (Score:1)
If they're going to name a robot after Isaac... (Score:1)
Why is walking so hard? (Score:2)
A question to any robotics experts out there... or heck, any mechanical engineer.
Why is walking so hard? It seems to me that the secret is having a balance sensor similar to the inner ear. If it has movable feet, then you use the balance sensor to keep the robot upright. Once you have that, then it seems like it's just a matter of kicking it forward. As long as it keeps one leg moving forward and keeps the other foot adjusting for balance, you should have natural walking motion.
Clearly it's not this easy, otherwise it would have been done long ago. What's wrong with this solution?
--
Re:Isn't this old news? (Score:1)
You can look a demonstration of Asimo in below link.
http://www.zdnet.co.jp/gamespot/gsnews/0011/20/
Sony's SDR-3X is impressive also. It is just 50 centimeter tall and 5 kilogram weight. And it can kick a soccerball and dance on one leg. You can look some pictures of SDR-3X in below link (but Japanese).
http://www.zdnet.co.jp/news/0011/21/sony.html
Why I think it's a hoax (Score:1)
But I've been downloading trans-oceanic quicktimes for the past couple of hours (at 28.8) and I still think it's a hoax.
Depictions of robots in sci-fi movies have always had the problem that they skip over the 'intermediate forms' between obviously-not-human and much-too-realistically human-with-a-few-small-flaws. And as far as I can see Honda's materials commit the same sin-- where are the intermediate forms???
A real example of an intermediate form [shadow.org.uk]
If you see a demo that in every detail resembles a person crouched in a special suit... what does Occam's razor say you should conclude?
Why did they need to give it a humanoid torso and arms? Why do these also move exactly like the human versions? Why is the head concealed in a visor, if not to give their actor a way to look around? Why do the P3 quicktimes [honda-p3.com] show only the cliched, anonymous box-with-blinking-lights inside, instead of showing off the real high-tech in the legs?
PROVE ME WRONG, PLEASE!
My page of robotics links [robotwisdom.com] if you're interested.
Robot (Score:3)
Hardware vs. clever algorithms vs. refined hacks. (Score:3)
Answers on a postcard please . . . :)
Unlimited Potential (Score:2)
Put some breasts on this thing, teach it a few new tricks, and we will have a winner!
Asimo? (Score:2)
Re:Priorities (Score:2)
But as for the "what happens when..?" question above, the answer is: we adapt. The implication is that we need to consider how it's going to affect us b4 we develop it. But we can't, and shouldn't, let that hold up developments which will improve the lives of everyone. The lobby groups were out in force when the car was invented - they reckoned it would spell the end of horses and everyone involved in that. And it did... but that doesn't mean we should have reconsidered allowing cars to be developed.
And BTW, the original cars were merely expensive, unreliable and slow toys. It took a couple of decades from the first working example b4 they were as fast as a horse, and another couple of decades b4 they were as affordable as a horse. And it took several years of work to get that first example working. Compare and contrast to the first human-size walking robot, which took several years to get working, and is still a slow, expensive toy...
Grab.
Re:Hardware vs. clever algorithms vs. refined hack (Score:2)
Well...
You're right, the ability to climb stairs with a bipedal robot is impressive, but when you find out they programmed it specifically for those steps it becomes a little bit less impressive.
The Honda engineers, in fact, admitted that if the steps were changed by only an inch in width the robot wouldn't be able to handle them. Instead of sensing it's environment there, it's just reverting to a subroutine for those steps. The thing that stinks is that they pull it off without anyone knowing. I might go so far as to say that it reeks.
But ya, the fact that they're walking on two legs in the first place is pretty cool. Now if only it didn't walk like a stick was up it's butt.
____
Not so new... (Score:2)
In general, these humanoid robots are NOT meant for physical research: building legs, as cool as it is, is not a new thing. Instead, these are research PLATFORMS, designed to test new theories in artificial cognition. The thinking, quite simply, is thus: if we want a robot to act like a human, it has to perceive the world (and interact with it) as a human does. Therefore, you equip it with sensors akin to ours (eyes, ears, touch), program the drivers and such (so that the robot knows it CAN see, CAN hear, CAN touch), and then you teach the hell out of it. Fun stuff!
Re:Why is walking so hard? (Score:3)
Aceleration/tilt sensors are used very often, sometimes a gyro, and there has even been talk of using an internal pendulum system for balance feedback. Using a system like this is a fairly good way of getting a walking biped. The problems arise when you don't have a level surface for your biped to walk on.
To perfectly mimic human walking, each leg would need 6 or more degrees of freedom. 2 for the hip, 1 for the knee, 3 for the ankle.
So imagine you have a bipedal robot that has a correct walk including hip movements for the transfer of balance between each leg. Now tilt the surface by a few degrees. Suddenly one leg needs to move less than the other, suddenly the ankle needs to change it's orientation so that the foot strikes the ground correctly, etc.
Now take this scenario to a rapidly changing surface such as a hill or rocky terrain. You can see how bad it could get.
But you would think that if you have a correct walking algorithm, no matter what surface you're on you would be able to adjust it for a good walking gait. Unfortunately, the robot needs some way of knowing what it will be stepping on before it ever does. It also needs to react quickly to feedback from sensors on it's feet and legs to adjust to anything it's previous scan didn't pick up correctly.
Because of the many factors, digital walking still has to get to the point where the computer can do all of this on the fly.
I think right now analog walking is the way to go. Analog creatures already walk quickly and easily on 4 legs, and it's easier to adapt them to two legs. It's just waiting on the right kind of sensors. My 4 legged bots are doing well, but I don't have enough money to work on a biped right now.
Oh well... hope this might help.
____
Honda (Score:2)
They're impressive beasts, but one has to wonder why they've spent nearly a billion dollars on this. For improving mass production techniques? Possibly, but there are surely better ways. As transport of the future? Hardly.
Sony too. (Score:3)
In an article [yahoo.com] off yahoo [yahoo.com] it appears that Sony too has a walking bot.
Re:Why I think it's a hoax (Score:2)
The real hoax, I think, is that this is a mechanical engineering achievement, not a leap forward in robotics -- notice how smooth the stage was? Adjust the angle of the stairs or put a bump in the floor and watch the thing fall on its ass.
That's not real robotics if it can't handle the real world.
Re:Isn't this old news? (Score:2)
No, that was an earlier version: the P3. This is just the latest in a line of Honda robots starting with the P2. The Slashdot article [slashdot.org] you refer to is concerned with the P3.
It's not terribly clear, probably because it's in Japan, but they are different things.