Shop OBEX P1 Docs P2 Docs Learn Events
Upbeat Futurists — Parallax Forums

Upbeat Futurists

ercoerco Posts: 20,256
edited 2011-06-16 22:17 in Robotics
From http://www.roboticstrends.com/consumer_education/article/student_robots_fly_roll_talk_and_learn_to_be_useful

"Someday you may have a personal robot to help around the house. It will move smoothly from room to room, avoiding obstacles, people and pets. It will pack your suitcases, clear the table and do the dishes. It might even be programmed to assign chores to the kids or wait for the cable guy."

Yeah, yeah. I love robots too. But this line sounds stunningly like something out of a Popular Mechanics magazine in 1955. It's just amazing to me that 50 years later, the mundane tasks listed are nowhere close to being solved. We have focused on solving some interesting technical problems: there are balancing robots, telepresence, and robots flying quadrotors with facial recognition. But Roomba and other robot vacuums are about as close as we've gotten to a life of luxury, with robots doing all the drudgery.

Wassup wit dat? :)

How we gonna fix it ? Keep adding Propellers to a robot until it gets smart enough to do something useful, but not smart enough to take over?

Who's with me?

Comments

  • Matt GillilandMatt Gilliland Posts: 1,406
    edited 2011-06-10 09:16
    erco - so you think we're moving to slowly?

    This is may be our future if we move too fast ;-)
    -Matt
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-10 09:47
    erco wrote: »
    From http://www.roboticstrends.com/consumer_education/article/student_robots_fly_roll_talk_and_learn_to_be_useful
    "Someday you may have a personal robot to help around the house. It will move smoothly from room to room, avoiding obstacles, people and pets. It will pack your suitcases, clear the table and do the dishes. It might even be programmed to assign chores to the kids or wait for the cable guy."

    Yeah, yeah. I love robots too. But this line sounds stunningly like something out of a Popular Mechanics magazine in 1955. It's just amazing to me that 50 years later, the mundane tasks listed are nowhere close to being solved. We have focused on solving some interesting technical problems: there are balancing robots, telepresence, and robots flying quadrotors with facial recognition. But Roomba and other robot vacuums are about as close as we've gotten to a life of luxury, with robots doing all the drudgery. Wassup wit dat? :) How we gonna fix it ? Keep adding Propellers to a robot until it gets smart enough to do something useful, but not smart enough to take over? Who's with me?

    Erco, your glasses are too Americanized!

    In Asian humanoid robot technology, the world is quite different and the roof is off the building and massive technology is unleashed - as the governments are generously financing home and service humanoid robots that are absolutely unbelievable in what they can accomplish and every month it's something new. Already these robots can cook the food, handle a skateboard, act as host, do the dishes, walk the stairs, operate construction machinery, entertain, roller skate, carry people in nursing homes, sing, play musical instruments like the piano and trumpet, ride a bike, dance Hip Hop, manage your internet, play soccer, play with children, do Tai Chi and Martial Arts, fight, deliver goods in hospitals, give lectures, open ceremonies, recognize people, and so on...
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-10 09:48
    We live in a special robotics era where humanoid robots are made to be specialized, usually performing one specific skilled task. In the next humanoid robotics era, the tasks will be combined.
  • ercoerco Posts: 20,256
    edited 2011-06-10 09:52
    @Matt: Yep, far too slowly for my tastes:

    It will move smoothly from room to room, avoiding obstacles, people and pets.

    We haven't really achieved that 100%. Even with 10 Pings! :)
  • ercoerco Posts: 20,256
    edited 2011-06-10 10:01
    @Humanoido. Those are all good, but specialized and isolated parlor tricks. Not unlike Elektro the Robot from 1937.

    From http://en.wikipedia.org/wiki/Elektro :

    Elektro is the nickname of a robot built by the Pittsburgh-based Westinghouse Electric Corporation in its Mansfield, Ohio facility between 1937 and 1938. Seven feet tall, weighing 265 pounds, humanoid in appearance, he could walk by voice command, speak about 700 words (using a 78-rpm record player), smoke cigarettes, blow up balloons, and move his head and arms. Elektro's body consisted of a steel gear, cam and motor skeleton covered by an aluminum skin. His photoelectric "eyes" could distinguish red and green light. He was on exhibit at the 1939 New York World's Fair and reappeared at that fair in 1940, with "Sparko", a robot dog that could bark, sit, and beg.

    Elektro, Asimo, Toyota's trumpet robot, Kawada HRP & Promet, all give choreographed performances to audiences. Amazing to watch, but not particularly flexible, and certainly "not ready for prime time".

    Did I actually say "parlor tricks"? I must be feeling fiesty (for a refreshing change). :)
  • WhitWhit Posts: 4,191
    edited 2011-06-10 14:38
    @Matt - excellent!

    @erco - I think I am with you - I find the sorts of things that Asimo does very frustrating. It is like they are pretnending that we can do what we really can't. I want to get there too, but it seems we are working in that directions - it is just harder than anyone ever imagined. Now - the fact that it is hard shouldn't stop us from trying. We just need to work harder. One foot in front of the other....
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-11 07:43
    erco wrote: »
    @Humanoido. Those are all good, but specialized and isolated parlor tricks. Not unlike Elektro the Robot from 1937.

    From http://en.wikipedia.org/wiki/Elektro :

    Elektro is the nickname of a robot built by the Pittsburgh-based Westinghouse Electric Corporation in its Mansfield, Ohio facility between 1937 and 1938. Seven feet tall, weighing 265 pounds, humanoid in appearance, he could walk by voice command, speak about 700 words (using a 78-rpm record player), smoke cigarettes, blow up balloons, and move his head and arms. Elektro's body consisted of a steel gear, cam and motor skeleton covered by an aluminum skin. His photoelectric "eyes" could distinguish red and green light. He was on exhibit at the 1939 New York World's Fair and reappeared at that fair in 1940, with "Sparko", a robot dog that could bark, sit, and beg.

    Elektro, Asimo, Toyota's trumpet robot, Kawada HRP & Promet, all give choreographed performances to audiences. Amazing to watch, but not particularly flexible, and certainly "not ready for prime time".

    Did I actually say "parlor tricks"? I must be feeling fiesty (for a refreshing change). :)
    I prefer to call it machine specialization. The attitude of cheap parlor tricks has prevailed even in the AI world. No matter what the robot can do or how sophisticated it becomes, someone, somewhere, somehow will always accuse it of being a cheap parlor trick. I think that's a rather prejudiced attitude towards a new machine life form that's beginning to emerge in the world, especially coming from an educated human man that specializes in robotics technology, has extensive education, and millions of years of evolution.

    Not sure about the problems from smoking incurred by Elektra, but it's reported, in the 1960s his head was given to a retiring Westinghouse engineer and his body was sold for scrap. I think humans can donate their bodies to science and donate body parts. Scrap is probably not the best word for this..

    http://en.wikipedia.org/wiki/Elektro
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-11 08:00
    Whit wrote: »
    @Matt - excellent!

    @erco - I think I am with you - I find the sorts of things that Asimo does very frustrating. It is like they are pretnending that we can do what we really can't. I want to get there too, but it seems we are working in that directions - it is just harder than anyone ever imagined. Now - the fact that it is hard shouldn't stop us from trying. We just need to work harder. One foot in front of the other....

    I understand the great difficulties but consider a hundred years is just a very small time of evolution in the overall time frame.

    ASIMO is but a child in the robotics world learning to walk, run and perform numerous tasks. I like to think of the decade from 2000 to 2010 as one in which humanoid robots began perfecting their physical attributes. You have to admit, walking up or down stairs is a remarkable achievement. As you say, one step at a time.

    In the 1980s there was a great robotics uprising, ushered in with the World's 1st Personal Robotics Conference held in Albuquerque New Mexico, but it died out when the public repeatedly asked of the robot's creators, "what can it do?" Robots like B.O.B. Brains on Board, rolled autonomously out on center stage, gave a speech, demo'd its mobility and lights, and the audience was unimpressed. "Now what can it do?"

    At least now, we have entered an age, 2010-2020 where humanoid robots can do more advanced things, albeit specialization, but they are rapidly advancing.

    Whit & Erco, I know you are frustrated like the rest of us, but just hang in there and in a few decades you'll get everything you wished for. Just take a look at the humanoid robotics world cup goals for 2030 and beyond to win over the human team.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-06-11 09:11
    Don't look at the news story, or the trend, as about robotics, per se. It's about using robots to demonstrate a panoply of cutting-edge technologies. Demonstrating these technologies using a robot as a host makes the technologies sexier, and more interesting.

    Anyone who can program a totally autonomous GPS-aware quadcopter with a face recognition camera (hint at the challenges: small camera, small processor, light weight) will easily command a $120K+ engineering job. That job probably won't have anything to do with robots.

    Microsoft has only recently discovered their Kinect technology is worth billions. When the API is formally released (maybe it has by now) lots of people will use it in robotics, and that's a great demonstration of the technology. But the real money is in the hundreds of other uses -- most of them haven't even thought of yet.
  • ElectricAyeElectricAye Posts: 4,561
    edited 2011-06-12 08:08
    Humanoido wrote: »
    ...

    In Asian humanoid robot technology, the world is quite different and the roof is off the building and massive technology is unleashed...

    Sounds like Fukushima.
  • ercoerco Posts: 20,256
    edited 2011-06-12 13:32
    @Gordon: Nice Gemini in your pic ! http://www.theoldrobots.com/gemini.html

    I bet your house is even fuller of robots than mine is!
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-12 15:55
    Originally Posted by Humanoido ...In Asian humanoid robot technology, the world is quite different and the roof is off the building and massive technology is unleashed...
    Sounds like Fukushima.

    Is that some kind of sick joke?
  • ElectricAyeElectricAye Posts: 4,561
    edited 2011-06-12 16:27
    Humanoido wrote: »
    ...

    Is that some kind of sick joke?

    Not exactly. In one of your previous posts above you stated the following:
    Humanoido wrote: »
    Erco, your glasses are too Americanized!

    In Asian humanoid robot technology, the world is quite different and the roof is off the building and massive technology is unleashed - as the governments are generously financing home and service humanoid robots that are absolutely unbelievable in what they can accomplish and every month it's something new. Already these robots can cook the food, handle a skateboard, act as host, do the dishes, walk the stairs, operate construction machinery, entertain, roller skate, carry people in nursing homes, sing, play musical instruments like the piano and trumpet, ride a bike, dance Hip Hop, manage your internet, play soccer, play with children, do Tai Chi and Martial Arts, fight, deliver goods in hospitals, give lectures, open ceremonies, recognize people, and so on...

    So I was just wondering why all that wonderful technology was not being applied to handling the situation in Fukushima. Instead, last I heard, they are planning on using old people because their cells do not replicate so fast:

    http://articles.cnn.com/2011-05-31/world/japan.nuclear.suicide_1_nuclear-plant-seniors-group-nuclear-crisis?_s=PM:WORLD
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-12 18:26
    So I was just wondering why all that wonderful technology was not being applied to handling the situation in Fukushima.

    As Japanese technicians stated, with redundant heavy intensive safeguards in place, they never anticipated a catastrophic failure of this magnitude, a double hit with a maximum earthquake and maximum tsunami - and didn't see the need to program humanoid robots for these intensive aftermath purposes. IMO such robots are ideally suited for nuclear waste cleanup and disaster management and more attention and detail should be given to this field. However, much of humanoid technology being developed is motivated towards elderly care. Those poor humans that gave up their lives at the nuclear power plant so that others could live should be greatly honored along with their families and I hope in the future it won't be necessary for humans to die over this.
  • ElectricAyeElectricAye Posts: 4,561
    edited 2011-06-12 19:07
    Humanoido wrote: »
    As Japanese technicians stated, with redundant heavy intensive safeguards in place, they never anticipated a catastrophic failure of this magnitude...

    I must admit, I was a bit shocked and puzzled that the Japanese didn't already have some kind of robot they could've used to pull fire hoses into those nuclear buildings. Here in the US, it seems not a week goes by without seeing a newfangled Japanese robot on the news, so the lack of robotics where and when it was really needed seemed almost irresponsible.
  • HumanoidoHumanoido Posts: 5,770
    edited 2011-06-12 22:56
    I must admit, I was a bit shocked and puzzled that the Japanese didn't already have some kind of robot they could've used to pull fire hoses into those nuclear buildings. Here in the US, it seems not a week goes by without seeing a newfangled Japanese robot on the news, so the lack of robotics where and when it was really needed seemed almost irresponsible.

    The Japanese are not an irresponsible people. Quite the contrary. But what forces result in the lack of robotics in a nuclear plant? Does the USA have radiation-hardened humanoids on standby in their reactor power plants? How about the Soviet Union? Or the UK?

    This is a good point to consider. I think the answer remains in the fact that each country is politically motivated in some unique way. In Japan, it deals with the elderly boom and the lack of caretakers. In the United States it deals with the military. In Korea, it's the prize money. For some countries it's a political statement.
  • prof_brainoprof_braino Posts: 4,313
    edited 2011-06-13 07:25
    erco wrote: »
    ... with robots doing all the drudgery...How we gonna fix it ?
    ... smart enough to do something useful, but not smart enough to take over?

    This nitch is already filled; in the US they are traditionally called "immigrants", and as long as they are cheaper than mechanicals, the scifi robots of the 50's are still a ways off. And its already too late, they've already taken over, just look at the Gubenanator; this is another American tradition.

    But I'll still help with the machines.
  • Martin_HMartin_H Posts: 4,051
    edited 2011-06-13 09:29
    Low labor costs are probably the biggest reason for lack of investment in robots to replace humans in unstructured situations. Basically real intelligence can be made by unskilled labor, while artificial intelligence requires skilled labor, so real intelligence has a cost advantage.

    Pick and place robots work in assembly lines because the environment is structured and the tasks are repetitive.

    Unmanned aerial drones can replace human pilots, but air is mostly empty space so again the environment is fairly predictable.

    Obviously dangerous environments are an exception because it increases the costs, but a radio controlled robot is likely to be more reliable.
  • ercoerco Posts: 20,256
    edited 2011-06-16 22:17
Sign In or Register to comment.