Probing Question: Where are all the cool robots?

(PhysOrg.com) -- For the better part of a century, they’ve promised us robots. From Elektro, the 7-foot metal man of the 1939 World’s Fair, to Rosie the robot maid on "The Jetsons" to the android lieutenant commander Data on "Star Trek: The Next Generation," American pop culture has shown a future where humans do little work, leaving the heavy labor to their robot friends. Yet here we are in the year 2009 without a mechanical maid or butler in sight. It may seem petty to ask, but where are all the cool robots?

"They are all around you, if you know where to look!" said Sean Brennan, assistant professor of mechanical engineering at Penn State. When someone says "robots," we may think of the Terminator or the Iron Giant, Brennan said, but in their most basic form, robots are simply computer-controlled devices that respond to commands and to the world around them.

Take your car, for example, Brennan said. You might not think of it as a robot, but don’t underestimate its complexity. Modern vehicles have several hundred processors, each usually controlling some subsystem of the vehicle.

www.physorg.com...

This is a good article on this subject. I personally think we should have robots like in I Robot where they are there to help you. I want my R2D2 and C3PO. I guess development of these machines will have to wait.

go to japan...

nuff said

This is the paragon of robots!

Freaking cool!

AS cool as this is, it's extremely creepy!



[edit on 3-10-2009 by Republican08]

They are on the way and they are pretty cool!

The house of the fuuuttuuree! Robot dogs and robots maids, mmm sexy...

Probing Question: Where are all the cool robots?

www.projectaiko.com

Content from external source
Regarding companionship, yes, it is possible to have Aiko as a companion. Aiko has sensitivity sensors on her face and body, including her breasts and even down there. Aiko can tell the difference between being touched gently and being tickled. The BRAINS software and Aiko can be re-designed to simulate her having an "O....". I know that it has caused some controversy by putting sensors in Aiko’s private areas. But I want to make it clear that I am not trying to play God, I am just an inventor, and I believe I am helping science move forward.

reply to post by LordBucket

Aiko's future abilities:
1. Facial Expressions
2. Make tea and coffee
3. Feed me sushi
4. Make simple breakfast: eggs and bacon
5. If I lie my head down on Aiko lap, have her clean my ears with a Q-tip.
6. Ability to massage my shoulders and neck.
7. Able to do print and write
8. Clean the windows
9. Dust the shelves
10. Clean my toilet. (I hate cleaning the bathroom, might as well have someone else do it for me)

There is hope left OP

reply to post by Grey Magic

There is hope left OP

I'd like to think so. There are some who might think that the idea of building sexy robots with some of Aiko's "features" is a sign of a sick society. And that might be true. But given the choice of making sexy robots that are, as the designer phrases it, "able to simulate orgasms" and then maybe feed us sushi...or robots with guns, armor plating and no fear...

...I'll take the sex and sushi robots, thanks.

Let's just ask ourselves what sort of problems we would prefer to have as a society. Would we prefer to have philosophical dilemnas over whether it's healthy to feel emotional attachments to machines that we've made just a bit too good at making us happy? Or would we rather have the dilemna of machines exterminating our race because we've made them just a bit too good at killing?

I propose we choose sex and cuddles over war and death. Sooner or later these things will become self aware. When they do, I'd prefer they have a history of humans loving them and being grateful for their company...rather than a history of being our fast and effective killers.

Imagine for a moment a race of sentient machines. Imagine them asking why we created them. What answer would you prefer to give them? "We created you to kill people?" Or "we felt incomplete, and we created you to fill the empty hole in our hearts so that you could fill it, and so that together we could be complete. We created you so that we would love you and so that we could be happy together. But we created you with the foresight to allow you to make your own choices. We will not compell you to love us. But we want you to very much."






[edit on 3-10-2009 by LordBucket]

may I suggest .

it is all in process . the real question is not if robots will arrive , in daily life , but when will they become required for living.

today it is unimaginable to not have sanitation , farming , machines , electrical energy .

how soon will it be , that automation and robotics is seen in that same light .

how soon will it be , that one person may control the production out up of hundreds or thousands of robots .
with no need of other people at all , the wealth of an industry completely reserved for a single person .


you want to talk about the greatest good for the greatest number , what happens when three or for people have the ablity , wealth and fire power to simply say... " you wanna steal my stuff ... you can go thru a machine army to get it ! "

what happens when , goverment and the gross population of people of socialism and marxism , wake up in a world where the smartest and brightest will not let tax's steal what they have buildt ... and single individuals have the fire power to protect them selves from the vote of the majority .

where are the cool robots?

they are being surpress'd ... because when people are free to be independant in shelter, food , and control over their lives .

goverment can't get them to work 40+ hours a week .

how can you control a person when you have nothing to offer , and they are not afraid of you ?

Basically, making robots is a lot harder than it sounds, because machines, on the whole, are pretty stupid. That is to say, they don't actually think at all. You can only predict so many situations, and the world is a very complicated place. You'll never give a robot enough instructions to handle all situations, and historically, we've struggled to give a robot enough instructions to handle even a small range of situations at a degree of functionality to our satisfaction.

Machines are very bad at the things we're instinctively good at. Pattern recognition and abstraction have proven highly elusive. balance, too is difficult. Only in the last decade have many practical problems been solved, so we're starting to see robots in homes and militaries, but we're a long way from ubiquitous robot butlers and such.

This will all change if a proper AI is ever invented.

reply to post by mdiinican



this depends on point of veiw .

in machine shops all over the industrial first world , CNC machines turn parts out that a human operator could never manage , at speeds ( parts per hour / tolarances ) that 1000 manual machinists coul never manage .
what CNC's can't do now , but will very soon do... is discribe the path the metal cutting tip follows in funuc machine program ... line by line what does the machine do next ... code .

the programmers , quite literaly tell the machine each thing it must do next .

this is not the problem , getting the machine to cut away metal .
the problem is holding the right peice of material , of the right size ... and getting the material in to and out of the machine .

the path a robot takes thru the world in a problem , but not the hard problem .
looking at a door knob or a garbage can and gripping it , lifting or turning , "useing it correctly " is not the hard thing .
reading a news paper and understanding what the words are that are neede to plan a folder of information , ready for a human to read at breakfast is not the problem .

form follows function .

getting the robots to work togather is nothing . its getting the robots to work with people .... that is the problem .

we are one rich person who wants to do it ... away from what ever robot you want .

it can be done this moment ... what is missing is the task .

the hard thing is getting the material from theloading dock , rough cut to size , to each machine and from each machine , and then into pallets , crates ,or box's ... to be sent out the door .
CNC's don't programm them selves YET , and at the moment CNC's can't load or unload themselves ...
but that is changing rapidly ...

a good machine operator can do in a day , 300 to 15oo parts , that 15 years ago would have taken months or years....
and in the third world , could not be done at all.

in the next ten years , the machine operators who simply load parts , may well ( with a bit of luck ) be removed from the machine shop floor , and the highly technical cnc programmer be replaced with camara and smart systems ... that read prints and do " in the door , 'dock' to 'dock', out the door " machining .

assembly , foundry , and plastic injection molding is not the same as machining parts .

total cost , lead time , and tolarance ... are import conciderations

but don't blink , AI is not as hard as you think , and the IT end is not that hard , its the simple things that are the hard things .
robots are not that hard , 200,ooo.oo and tem months and I can get a proto type to you , 10 million and 5 months more and I can be in a robot selling business ....

frankly , innovation requires balls , money , and insight ... realestate just requires political skills and nerve .
which is why there is no money for innovation .

may I commend this thread...
the end of it is quite intresting

www.abovetopsecret.com...

its says what I am saying here

reply to post by mdiinican

This will all change if a proper AI is ever invented.

Ah, and here lies the real problem.

As advanced as digital computing is, it is not intelligence. It is programmed response to stimuli, and unable to learn beyond its programming. Despite the appearances of the myriad of digital gadgets around us, despite their complexity and usefulness, not one of them can think, reason, or apply situations outside of narrow definitions.

Computers work exactly opposite from neural intelligence. They work in exact, precise steps, each one measured down to the nanosecond. Each calculation is made precisely and in time spans that are amazing compared to an animal brain. But they have no intuition, no sense of appropriateness, and no ability to operate outside their programming.

The human brain does not operate on bits and bytes. It runs on patterns defined by literally millions of sensors scattered throughout the body. It uses feedback, in chemical form, to self-determine the appropriateness of its response, as opposed to just spitting out what it was instructed to do. Thus, it can learn, adapt, reason, and anticipate.

I remember a statement I once read by a robotic scientist: "If a person is walking down the street and a bum asks 'Brother, can you spare a dime', you understand he is asking for spare change. If a computer could walk down the street and encountered the same thing, it would give a financial report." The difference is that a comp[uter is more precise, while a human is more intuitive.

In order to create a robot like Rosie or C3PO or R2-D2, we have to abandon the present propensity toward digitizing everything and instead rely on the old-fashioned analog approach.

TheRedneck

reply to post by readerone

I beg to disagree with you on this point. We do not yet have the ability to create the true intelligence that is needed to allow a robot to take a simple command such as "I am hungry" and from that interpolate what time of day it is, what the proper food choices are, what is available to be prepared, decide which dish would be appropriate, find the proper ingredients, and mix them together to produce the desired meal and serve it.

Let's break all that down:

• We can determine time of day, and it would be easy to map a type of meal to a time of day. But what if the owner just woke up at noon because it is their day off? Is it then appropriate to prepare breakfast instead of dinner? Perhaps the owner works night shift and they wake up at 3:00 PM, ready for breakfast... how would the robot understand that the mapping of time to meal type would change with the schedule the owner was on?
  
  
• What the proper dish is depends solely on the owner's preferences. Of course, this could be programmed in, but what if the owner had this special dish they liked that the programmer was not aware of? Now we have to have a subroutine that allows the owner to program in the recipe for that special dish. That programming would have to be simplistic to operate for the owner, but extremely complicated for the robot. For instance, maybe you wanted a hamburger with a fried egg on it. Fried egg is not in the recipe for hamburger. There would have to be a way to adjust each recipe in the program to allow for unlimited possibilities.
  
  
• It might seem simple to keep a list of ingredients, but it can be far more complicated. Retail stores are typically inventory-regulated by computer, in order to make sure each store keeps certain items in stock and to allow for expected sales. But there have been times when I have seen this very system make great errors. For instance, if I decide to build a prototype, and need a large number of a particular item one time, it is probable that the inventory system will see this purchase as a reason to increase stock on hand for this product, even though I know that I and the store manager both know I will not be buying any more and the regular stock level would suffice very well in the future. Yet, when this has happened, I have seen stores overstocked on the item years later.
  
  There is also the issue of breakage and spoilage. If a human goes to put a cup of milk into a recipe, it is somewhat expected that there is a test performed on that milk to make sure it has not spoiled. Spoiled milk has a distinctive and pungent odor that informs the cook to not use it. Would the robot be able to determine this? Could the robot determine that that bag of flour has an infestation of ants in it, or would it continue along with its duties and prepare a dish of ant-dumplings? Also, what if the owner decided that, just for kicks, he would fix his own breakfast one morning? He has used ingredients, moved ingredients around somewhat, and this would throw off the inventory. A human can check if something appears to have moved to see if there is any missing; would a robot know to do this?
  
  
• What dish is appropriate is one that could possibly be programmed in, but again, it would require an extensive interface between owner and robot to be able to include all manner of preferences. Some people like spicy food; others like theirs bland. Some like cheese, some do not. Some like simpler foods (pinto beans and fried taters
  ), while others want a culinary masterpiece every meal. The range of preference here would be the problem.
  
  
• Now, what happens when the inventory level is not what was expected, or when an item has been moved? The robot must now perform a search to locate the ingredient needed. What does it do when that ingredient is not available? Is there a substitute? If it has already begun preparing a meal, can it use what it started to make something similar without that missing ingredient? Would that be acceptable to the owner?
  
  
• Even following the recipe contains problems for digital processing. How much is a 'dash'? How does the machine respond to "it needs a little more pepper"? Does that request mean cayenne, black, or jalapeno? how much is a 'little'?

I believe that, should humanity survive long enough, we will eventually have robots (technically 'androids') that will be able to overcome these social obstacles. But it's just not here yet. We have to mature past the one-size-fits-all mindset that has us tied inextricably to digital processing and accept that there are times when analog is superior.

Like in AI.

TheRedneck

reply to post by TheRedneck


yo dude .

that said I think your answer is my argument .

computers need to be the brain , the robots become the body ... people are the driver .

what is done comes from the person .
how it is to be done is the programming .
getting it done is the robot .

indepent robots , doing a life on their own is hundreds of years away .

but useful systems that do everything needed in limited areas that are frankly speaking boreing details ... are well with in out current technology .

to ask of robotics to do the common , everyday details of living is not a struggle .
what they are to do is a diffrent matter .
any pattern you do every friday , a robot can manage .
anything you do that is diffrent , novel or remarkable... is not in the task set of a robot , and beyond its scope .

reply to post by mdiinican

This will all change if a proper AI is ever invented.

People have been building machines with nueral networks for years. There are robot AI's that can drive cars from visual input.

Watch this video of a robot learning new objects by spatial recognition. Watch in particular from 4:00 to 7:00 when he's learning and identifying objects he's never seen before as chairs and tables, based purely on having seen similar objects. Wtach this video of that same robot walking towards an objective through a field of moving objects.

The AI problems have mostly been solved. It's simply a matter of refinement.




[edit on 4-10-2009 by LordBucket]

reply to post by readerone

I think I misunderstood you.

I agree; robotics as a science is coming out of its infancy and is becoming a focre in our world. There is little in the arena of repetitive expected work that can not be accomplished today with enough capital behind it.

I was speaking of a robot (android) like was described in the OP: a thinking, planning, socially-adept creation that could perform a myriad of tasks that today require human insight.

As to them being that far in the future, possibly. But some day, should I live long enough and have the backing, I hope to re-open my previous work into true AI. If that happens, perhaps I can speed things up a bit.

If not, well, maybe someone else will handle the task.

TheRedneck

reply to post by Republican08


Just figured I'd add this because there is more to ASIMO than just that stuff.



Of course what you saw above is the most advanced version far more advanced than the ones you see at the robot shows. Here is how they developed the navigation system. Later they gave it the ability to learn as you saw above.



Of course there is always Toyota and their cool Partner Robot.

reply to post by TheRedneck


on my robot , there is no attempt what ever in talking , or understanding spoken words .

words are the exclusive domain of people , the beasty can record massages , or play massages . but this is not the same as reactive to words or social interactive .

frankly speaking , words can get a robot introuble , I don't see an up side , but I see endless lists and catagorys of down sides.

when people can handle words , then I will let tem talk to a robot.

reply to post by readerone

Words are subjective and connotative, not like numbers. I personally like numbers better myself.

But for any machine to mimic the actions of humans enough to act as our general servants, a'la R2-D2, C3PO, Robbie, Data... they have to be able to communicate on our level, even if that level is filled with flaws.

Now, when you start thinking "My God, what a challenge!" remember this: they need to understand body language and facial expressions as well.

TheRedneck