Humanoid robot learns how to run

pascal

Veteran
http://news.bbc.co.uk/2/hi/technology/4098201.stm
Car-maker Honda's humanoid robot Asimo has just got faster and smarter.
The Japanese firm is a leader in developing two-legged robots and the new, improved Asimo (Advanced Step in Innovative Mobility) can now run, find his way around obstacles as well as interact with people.

Great improvment. Now add some advanced vision control system, better voice recognition, better AI (edited) and voila :)

Has Asimo any relation to Asimov, or is just coincidence?

Some movies: http://world.honda.com/HDTV/ASIMO/
It makes a lot of noise, looks like a mixer :LOL:
 
Everytime i see those big things i just wonder when they'll be FAST.
I mean, we humans are not exactly Flash, but we can be very very fast when it comes to reflexes, change of direction, everything really...
I wonder when they'll be able to make robots that are as quick as us... Although i wouldn't want to know what could happen when that day comes *thinks about Terminator/iRobot/whatever apocalypse scenes"
 
better IA

better AI .. perhaps??

I wonder when they'll be able to make robots that are as quick as us... Although i wouldn't want to know what could happen when that day comes *thinks about Terminator/iRobot/whatever apocalypse scenes"

The year 2035/6 .. at least according to some people. Actually I think it 2036.

US
 
Unknown Soldier said:
better IA

better AI .. perhaps??

I wonder when they'll be able to make robots that are as quick as us... Although i wouldn't want to know what could happen when that day comes *thinks about Terminator/iRobot/whatever apocalypse scenes"

The year 2035/6 .. at least according to some people. Actually I think it 2036.

US

Well, the war should have started in 1997 right? That's what Sarah Connor said...
 
I was actually referring to the "Singularity". Apparently people believe that, by 2036, computers/robots will create other robots and at this time the first true AI enhanced robot will come to "life".

http://singinst.org/
http://www.kurzweilai.net/articles/art0133.html?printable=1

There's all different dates though so not 100% sure.

Got another quote saying 2026.

The idea that human progress would reach a "singularity" originated in Doomsday: Friday, 13 November, A.D. 2026, Science 132, 1291-1295 (1960) by von Foerster, H, Mora, M. P., and Amiot, L. W. The mathematical singularity appeared in that paper's human population model (Doomsday equation). Von Foerster argued that human's abilities to construct societies, civilizations and technologies do not result in self inhibition. Rather, societies' success varies directly with population size. von Foerster found that this model fit some 25 data points from the birth of Christ to 1958 with only 7% of the variance left unexplained. Several follow-up letters (1961, 1962, …) were published in Science showing that von Foerster's equation was still on track. The data continued to fit up until 1973. The most remarkable thing about von Foerster's model was it predicted that the human population would reach infinity or a mathematical singularity, on Friday the November 13th, 2026.

Technological singularity is a term with multiple related, but conceptually distinct, definitions. One definition has the Singularity as a time at which technological progress accelerates beyond the ability of current-day human beings to understand it. Another defines the Singularity as the culmination of some telescoping process of accelerating computation taking place in this universe since the beginning of human civilization or even life on Earth. Yet another defines the Singularity as the emergence of smarter-than-human intelligence, and subsequent cascading consequences that are not possible to predict or, perhaps, guide or even influence.

http://www.iridis.com/glivar/Technological_singularity

You know .. i've just noticed that they say that the Honda robot has increased it's speed to 3km/h .. uuhh duuhh!!! You made it bigger than it was beofre .. of course it's gonna go faster.

The "run" he is now capable of is perhaps not quite up to Olympic star Kelly Holmes' standard. At 3km/h, it is closer to a leisurely jog.

Its makers claim that it is almost four times as fast as Sony's Qrio, which became the first robot to run last year.

The criteria for running robots is defined by engineers as having both feet off the ground between strides.

Asimo has improved in other ways too, increasing his walking speed, from 1.6km/h to 2.5km, growing 10cm to 130cm and putting on 2kg in weight.
 
london-boy said:
Everytime i see those big things i just wonder when they'll be FAST.
I mean, we humans are not exactly Flash, but we can be very very fast when it comes to reflexes, change of direction, everything really...
I wonder when they'll be able to make robots that are as quick as us... Although i wouldn't want to know what could happen when that day comes *thinks about Terminator/iRobot/whatever apocalypse scenes"

There are robots 100x faster than humans, but not humanoid - rather industrial robots doing stuff on the assembly lane or such.
 
BTW, I wholeheartedly recommend Vernor Vinge's "A Deepness In The Sky" and "A Fire Upon The Deep". Great books!

(that's the guy who first used the term "singularity" for an AI)
 
_xxx_ said:
london-boy said:
Everytime i see those big things i just wonder when they'll be FAST.
I mean, we humans are not exactly Flash, but we can be very very fast when it comes to reflexes, change of direction, everything really...
I wonder when they'll be able to make robots that are as quick as us... Although i wouldn't want to know what could happen when that day comes *thinks about Terminator/iRobot/whatever apocalypse scenes"

There are robots 100x faster than humans, but not humanoid - rather industrial robots doing stuff on the assembly lane or such.

U know what i mean!! ;) Faster at doing stuff.. like playing footbal, shagging... normal stuff.
 
Unknown Soldier said:
I was actually referring to the "Singularity". Apparently people believe that, by 2036, computers/robots will create other robots and at this time the first true AI enhanced robot will come to "life".
...
My guess the singularity will happen after 2036+10 because it is not enough to have the computational power, you need to know/learn how to use it (algorithms, etc..).
 
london-boy said:
U know what i mean!! ;) Faster at doing stuff.. like playing footbal, shagging... normal stuff.

Shagging EVEN FASTER? OMG! The girl/guy respectably wouldn't even notice than...
 
I see no reason to believe in singularity. I mean...computers get faster and faster, have more and more processing capacity, but still AI develops at incredible pace of a tetraplegic turtle.

People keep saying that in 20-30 years, computers will have the processing capacity of the brain, and the there will be true AI and even super-intelligence. My prediction, is that they will still be as dumb as a door knob.

My only hope in reaching true AI, will be by emulating the brain, through a reverse engineering. We're only now starting to actually learn how the brain works, so it's still hard to predict when this will happen. But IMO, it's the only way. I sure as hell don't believe that any true progress will be made using the weak-AI approach (which is what every researcher seems t be doing these days).

About super-inteligence, maybe it will happen once we identify what makes one person smarter then the other.
 
Alejux said:
I see no reason to believe in singularity. I mean...computers get faster and faster, have more and more processing capacity, but still AI develops at incredible pace of a tetraplegic turtle.

People keep saying that in 20-30 years, computers will have the processing capacity of the brain, and the there will be true AI and even super-intelligence. My prediction, is that they will still be as dumb as a door knob.

My only hope in reaching true AI, will be by emulating the brain, through a reverse engineering. We're only now starting to actually learn how the brain works, so it's still hard to predict when this will happen. But IMO, it's the only way. I sure as hell don't believe that any true progress will be made using the weak-AI approach (which is what every researcher seems t be doing these days).

About super-inteligence, maybe it will happen once we identify what makes one person smarter then the other.

A brain can't let you run a game or whatever, even if it would be a perfect replica. A brain can't calculate in teraflops. It will have to be something else.
 
I believe that the brain is what it is today because of millennia of evolution. Part of that evolution was made possible by the environment, experience, and most importantly our ability to "do things".

It will take a long time for AI to reach our level of intelligence, and i believe we do need to start from having a body, and developing the brain in it. Like we were first a body that eats and runs and procreates, then "got intelligent".
 
I agree with you to a certain extent. In order to delevop into an intelligent being, the artificial brain will need, like anyone else, to be constantly receiving sensory stimulus (vision, audio, tact, scent) and have a body (even if virtual) to interact with this environment. It will need to feel both pain and pleasure, and have to learn his way into avoiding the first and seeking the second. That IMO, is ours, and every other animals prime motivation in all things.

I think it would be a monumentaly hard (baring the impossible) task for us to create an artificial brain from scratch. Luckily, we don't have to. All we have to do is assimilate what nature has already produced in millions of years of evolution. That's where my faith in true AI lies.
 
Yes and to add something, that is why i don't think a "desktop PC" as such will ever be as intelligent as a human. It just can't. It doesn't have enough stimulae from the outside world. It might be able to "know" more stuff than us, it might be able to do maths faster than us, but we will still "win" in videogames, unless the AI cheats, to use a stupid example. For a long time.
 
The other way around, human brain is not capable of calculating really fast. There would be no point in creating a replica of a brain for the tasks usually done by a computer.
 
Back
Top