News:

Actually sport it is a narrative

Main Menu

Robotics

Started by Inevitable Droid, November 05, 2010, 12:33:41 PM

Previous topic - Next topic

Inevitable Droid

Any fans of robotics on this message board?  If so, I'll start posting links to articles from time to time.

I love these guys:

ASIMO - http://world.honda.com/ASIMO/

TOPIO - http://topio.tosy.com/about.shtml

Ultra Trencher 1 - http://www.engadget.com/2008/03/22/smd-ultra-trencher-1-starts-its-new-job-laying-pipes-and-cables/

Mars Rover - http://marsrovers.jpl.nasa.gov/home/index.html
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

karadan

Hell yeah! The subject of robotics and AI is of great interest to me. Thanks for the links. They will pass the time at work today, no doubt.
QuoteI find it mistifying that in this age of information, some people still deny the scientific history of our existence.

Asmodean

I've had enough trouble with the making of robotic systems that I, for one, will stay away from those links  :P
Quote from: Ecurb Noselrub on July 25, 2013, 08:18:52 PM
In Asmo's grey lump,
wrath and dark clouds gather force.
Luxembourg trembles.

Inevitable Droid

Robonaut - a humanoid robot designed to perform tasks that weren't specifically designed for robots to do; I.e., tasks humans would have done; and intended to be placed indefinitely on the International Space Station during the final mission of the Shuttle, which was recently postponed.
[youtube:1mq69d7m]http://www.youtube.com/watch?v=ZbYj10RYD8c[/youtube:1mq69d7m]
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Inevitable Droid

Leonardo - the most sophisticated social robot in the world today:

[youtube:3ahq03o6]http://www.youtube.com/watch?v=ilmDN2e_Flc[/youtube:3ahq03o6]
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Inevitable Droid

Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Achronos

Check this out, Skynet is coming along nicely.  :D

[youtube:1m9qaz1p]http://www.youtube.com/watch?v=cFVlzUAZkHY[/youtube:1m9qaz1p]
"Faith is to believe what you do not see; the reward of this faith is to see what you believe."
- St. Augustine

Faradaympp

Quote from: "Achronos"Check this out, Skynet is coming along nicely. :eek:. I would really appreciate it.
"It's ironic that a god who created intelligent beings would want their blind devotion."-Anonymous

CAUTION-Staring at burning bushes may cause blindness. ;)

Inevitable Droid

Quote from: "Faradaympp"Also I'm doing a project on the ethical, legal and social implications of an AI entity in our societies and I'm having a writers block as far as topics go. If anyone has any questions or suggestions for questions at all can you please post them :eek:. I would really appreciate it.

Here is the crucial question - What behaviors would a robot have to exhibit so as to compel us to accept it as a person rather than an instrument?  The next question follows logically.  Is there any reason to treat a robotic person differently from a biological one?
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Tom62

QuoteA Lego Mindstorms robotics kit controlled by an HTC Nexus One smartphone successfully untangled a Rubik's Cube puzzle in 12.5 seconds at this weeks ARM technical conference in Silicon Valley. The current 3x3x3 cube-solvers's 15-second average represents a substantial improvement over the 25-second solutions of an earlier version, which was powered by a circa-2006 Nokia N95 smartphone, thanks to a faster (1GHz) CPU, more RAM, and revamped cube-solving algorithms. ARM Engineer David Gilday, who created the robotic cube-solver, claims the current version's algorithms can handle cube complexities up to 100x100x100, assuming he build the mechanics. In terms of racing humans, Gilday says the Lego robotics kits can only manage around 1.5 moves per second, whereas human players can make between 5 and 6 moves per second, amazingly enough.

Source: http://deviceguru.com/android-phone-con ... biks-cube/
The universe never did make sense; I suspect it was built on government contract.
Robert A. Heinlein

Inevitable Droid

Quote from: "Tom62"Source: http://deviceguru.com/android-phone-con ... biks-cube/

Awe-inspiring!!  Watching those clips I was on the verge of tears!!  Thank you so much for posting this!! :hail:
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Ultima22689

Don't forget memristors, supposedly capable of synaptic logic, positronic brains for our robo bros and for us trans humanists to leave the biological shell behind are on the way. Memristors are very real and will be commerical in 3-5 years. HP is backing it fully and pouring buckets of money into research. We're entering a whole new paradigm of computing. The implications for robots, AI and human evolution can't be understated.

http://www.msnbc.msn.com/id/36605027

http://www.zdnet.com/blog/foremski/tha- ... uting/1309

Persimmon Hamster

Quote from: "Inevitable Droid"Any fans of robotics on this message board?
I'm a fan of artificial intelligence, including robotics.  I have a question that perhaps you can answer...

First let me say that, professionally, I am a software engineer with a bachelor's degree in computer engineering.  Also let me say that I follow news about robotics only inasmuch as it happens upon me through my normal channels -- normally I do not seek such news out as part of a regular routine.  And while I have made some meager efforts (sporadically through the years and of limited duration) to read up on fields relating to artificial intelligence, I would still consider myself effectively ignorant on the vast majority of related topics.

All of that said...sometimes I personally ponder, within the scope of my own ability to articulate and consider, what I would perhaps call "engineered consciousness", in preference to "artificial intelligence".  First, I substitute "consciousness" for "intelligence" because I believe we have already successfully achieved levels of intelligence, and I feel that should not be the final goal and this should be made clear in the terminology.  For example, when you design an algorithm to solve a Rubik's Cube, you have written code to model an intelligent thought process.  What I would be ultimately hopeful in (and indeed, a likely final destination for the field) is a level of intelligence such that it is indistinguishable (to us) from our own -- that is, what would appear to be a conscious entity.  An entity at the sight of which we could no longer claim some kind of monopoly on concepts of consciousness, self-awareness, free will, morality.  I substitute "engineered" for "artificial" because I find nothing artificial about any of it.  I regard anything in the universe that ever has or ever will exist or occur as having existed or occurred wholly naturally.  I feel the term "engineered" helps avoid confusion about this view in that it more clearly suggests simply that one being believing itself to be conscious has engineered another.

Now, let me get to my question (though another paragraph or two may yet intervene).  The efforts in the field which I do happen upon from time to time in the news generally appear to me as almost too interested in modeling human-or-animal-like intelligence, and almost too apparently insistent that achieving such would require total duplication of our senses and physical motor capabilities.  For example, many robots attempt to simulate vision, and to record and interpret such data [on a level similar to humans] they [would seem to] require an immense amount of space and highly complex supporting pattern recognition/analysis algorithms.  I acknowledge fully this may simply be the appearance that ends up filtered to me by the mass media which understandably would tend to focus on results more easily understood by the "layman" viewing audience (e.g., Kismet).  Personally I tend to think that a path to engineered consciousness which duplicates human senses, human motor ability, and human brain functions might well succeed, but might take a lot longer than we expect--even with the current projected trends of computer capability vs. cost being as they are.

So what I find myself wondering is, might an approach to achieving engineered consciousness entirely within software be able to succeed sooner?  Consider a program which has "sensory" inputs...perhaps in the form of an input data buffer for a digestive system and an output character buffer for a vocal system.  Consider a data array for an environment, and a coordinate based "motor"/location system within that environment.  Such an approach could simplify pattern analysis algorithms necessary to duplicate memory and functions of consciousness dependent thereon.  Add "hardwired" basic needs and "instincts" into the mix...create a sufficiently complex "environment"...add a simplified capability for learning...and observe, teach, watch, and wait for consciousness to emerge.

I believe what I am getting at here may most aptly be referred to as a "symbolic" approach to "artificial consciousness".  But I lack sufficient knowledge about the state of the field itself and the terms used within it to be certain.  My question, then, is this:

As possibly a more diligent student of [or contributor to] the fields of AI/AC/robotics than I, how would you most accurately categorize an approach such as I have suggested, and what do you know about the current state of the field with regard to such an approach...and how do you feel about such an approach?
[size=85]"If you wish to make an apple pie from scratch, you must first invent the universe."[/size]
[size=75]-- Carl Sagan[/size]

[size=65]No hamsters were harmed in the making of my avatar.[/size]

Inevitable Droid

Quote from: "Persimmon Hamster"I'm a fan of artificial intelligence, including robotics.

Me too.  That's the most accurate way to describe me, in this context.  I wish I were an expert in the field, but I'm not.  I merely admire those who are, and the products of their excellent and important work.  Roboticists are among my heroes.

QuoteFirst let me say that, professionally, I am a software engineer with a bachelor's degree in computer engineering.

I was a professional computer programmer for approximately five years in the 1980's.  I coded in BASIC and PL/I.

QuoteWhat I would be ultimately hopeful in (and indeed, a likely final destination for the field) is a level of intelligence such that it is indistinguishable (to us) from our own -- that is, what would appear to be a conscious entity.  An entity at the sight of which we could no longer claim some kind of monopoly on concepts of consciousness, self-awareness, free will, morality.

Barring nuclear or other Armageddon, I deem it highly likely that roboticists will eventually produce machines that give all, and I mean all, of the appearances of sapience.  Robots already have free will as I define it, namely, "the capacity to make decisions that aren't entirely dependent on external causes that are active at this present moment."  They don't yet have responsibility as I define it, namely, "the capacity to decide, and the capacity to carry out the decision, to modify one's own decision-making parameters."  But I deem it highly likely that if humanity avoids extinction long enough, robots will eventually exist that have responsibility as well.

Free will and responsibility are attributes of the decision-making apparatus.  I define psyche as, "multiple parallel processes, continually running, characterized as perceiving, thinking, emoting, lusting, and deciding."  Robots already exist that do all five, and I mean all five.  They don't do any of it to the level of sapience, but they approximate the level of the most primitive invertebrates, and in the area of thinking they achieve some idiot savant feats of brilliance, such as solving Rubik's cubes or winning games of chess.  But as far as any of us humans can plausibly speculate, robots lack wakefulness, which I won't define, because there aren't any English words that will define the term in anything but a circular manner.  I will merely have to trust you to understand what I mean when I say, robots do all the things a living creature with a brain does, but robots presumably aren't awake to the fact that they're doing any of these things.  The reason they (presumably) aren't awake to any of it is because humans haven't programmed wakefulness into the software, for the very good reason that humans can't begin to imagine what it would even mean to program wakefulness into something or anything.

On the day when a robot first says to its masters, "I am awake," I predict that its masters (1) won't know whether to believe the robot or not; (2) won't know how to test whether the robot's statement is true or not; (3) won't have any idea how that which the statement denotes could possibly have come to pass; and (4) will be faced with the moral dillemma of how to respond.  (Incidentally, the only way we will ever have any inkling that a robot might be awake, is if it tells us that it is.  Perceiving, thinking, emoting, lusting, and deciding can all be programmed and can all be deduced as occurring by observing behavior, but wakefulness to perceiving, wakefulness to thinking, wakefulness to the remaining three, will be utterly undetectable to any human sense or any technological monitoring device.)
 
Ultimately, the truth or falsehood of the robot's statement will be ascertained by humans subjectively, based on human emotion and/or human appetite.  Humans who hope the robot is awake will decide to take for granted that it is.  Humans who hope that it isn't will decide to take for granted that it isn't.  The first group will lobby for the robot to be treated morally and legally the same as humans.  The second group will lobby for the robot to continue to be treated morally and legally the same as a hammer or wrench.  The seeds of violent conflict will be planted.  Guns will be loaded and kept handy.  The problem with subjective truth is that sometimes it forces a line in the sand.  But at least in this case, the question at issue will actually matter.  The robot will clearly exist, and will clearly be awake or not, and if awake, will clearly have a claim on moral and legal recognition.
 
QuoteSo what I find myself wondering is, might an approach to achieving engineered consciousness entirely within software be able to succeed sooner?  Consider a program which has "sensory" inputs...perhaps in the form of an input data buffer for a digestive system and an output character buffer for a vocal system.  Consider a data array for an environment, and a coordinate based "motor"/location system within that environment.  Such an approach could simplify pattern analysis algorithms necessary to duplicate memory and functions of consciousness dependent thereon.  Add "hardwired" basic needs and "instincts" into the mix...create a sufficiently complex "environment"...add a simplified capability for learning...and observe, teach, watch, and wait for consciousness to emerge.

I see your point, I think.  The only thing a microprocessor actually experiences is an electrical charge that is either on or off; I.e., ones and zeroes, as humans choose to designate those on/off states.  Perhaps a rich environment of on/off states could be run through the microprocessor... patterns, electrical melodies.  The music would be one-dimensional, like listening to one flute played solo, but still, perhaps the robot might somehow become awake to the melody.  Perhaps the robot, when otherwise idle, could be left in music appreciation mode, in hopes that one day, for reasons no one will be able to explain, the robot might awaken.

I've never heard of anyone attempting anything along these lines.
Oppose Abraham.

[Missing image]

In the face of mystery, do science, not theology.

Persimmon Hamster

Quote from: "Inevitable Droid"I was a professional computer programmer for approximately five years in the 1980's.  I coded in BASIC and PL/I.
I suspected by your posts around the forum that you were no stranger to programming or that, if you were, you would be well suited to the task.  I've been coding professionally for coming up on 8 years (during that time, mainly with Java, C++, and C#) and prior to that as a hobbyist/student for another 8 (during that time, primarily QBasic and similar, C, and C++).

Quote from: "Inevitable Droid"On the day when a robot first says to its masters, "I am awake," I predict that its masters (1) won't know whether to believe the robot or not; (2) won't know how to test whether the robot's statement is true or not; (3) won't have any idea how that which the statement denotes could possibly have come to pass; and (4) will be faced with the moral dillemma of how to respond.  (Incidentally, the only way we will ever have any inkling that a robot might be awake, is if it tells us that it is.  Perceiving, thinking, emoting, lusting, and deciding can all be programmed and can all be deduced as occurring by observing behavior, but wakefulness to perceiving, wakefulness to thinking, wakefulness to the remaining three, will be utterly undetectable to any human sense or any technological monitoring device.)
I agree with #1, 2, and 4, but not necessarily 3.  On "wakefulness" (or consciousness), I remain skeptical that it is anything as special as most would like to believe, and might argue we really are incapable of discerning its presence/existence not only in potential future robots, but in other human beings and in ourselves as well.
 
Quote from: "Inevitable Droid"I see your point, I think.  The only thing a microprocessor actually experiences is an electrical charge that is either on or off; I.e., ones and zeroes, as humans choose to designate those on/off states.  Perhaps a rich environment of on/off states could be run through the microprocessor... patterns, electrical melodies.  The music would be one-dimensional, like listening to one flute played solo, but still, perhaps the robot might somehow become awake to the melody.  Perhaps the robot, when otherwise idle, could be left in music appreciation mode, in hopes that one day, for reasons no one will be able to explain, the robot might awaken.
Well, what I am proposing would not be static.  That is, the "environment" would not simply consist of data ordered in a particular manner and then left entirely alone.  As I imagine this approach, there would need to be other, external, algorithms in place to simulate regular activity in the environment external to the "engineered consciousness" itself.  It seems to me, so far, that external forces are key in driving consciousness.  There would also be external interaction with the programmer, perhaps putting food into its "mouth" before it could learn to find it and feed itself, and certainly putting words into its "ears" to teach it language -- or we'd have no way, ourselves, to perceive it state "I am awake".  Personally I would never propose consciousness could emerge in the absence of external forces.

Quote from: "Inevitable Droid"I've never heard of anyone attempting anything along these lines.
Fair enough.  In composing my original post above, I consulted Wikipedia articles on artificial intelligence and artificial consciousness.  It seems perhaps the most promising activity in the latter field may revolve around neural networks, a concept I only superficially understand and which I suspect to be different from what I am describing--what I am describing seems as though it would be considered a "symbolic" approach, but I could be wrong.

I'll leave you with an excerpt from the Wiki AC article which I found particularly intriguing, and which might just lead me to read the book that is mentioned:

Quote from: "Wikipedia"http://en.wikipedia.org/wiki/Artificial_consciousness
Igor Aleksander, emeritus professor of Neural Systems Engineering at Imperial College, has extensively researched artificial neural networks and claims in his book Impossible Minds: My neurons, My Consciousness that the principles for creating a conscious machine already exist but that it would take forty years to train such a machine to understand language.
[size=85]"If you wish to make an apple pie from scratch, you must first invent the universe."[/size]
[size=75]-- Carl Sagan[/size]

[size=65]No hamsters were harmed in the making of my avatar.[/size]