Slashdot Mirror
Learning Robots
carnun writes "The NewScientist is reporting on a new snakelike robot that uses genetic algorithms to change its motion pattern so that it can still function after sustaining damage." Roland Piquepaille sends in a report about another "learning" robot, named Adam, which is designed to seek out and feed from "flowers".
Again, I think it looks like it is time for some robot insurance!
"A synonym is a word you use when you can't spell the word you first thought of." - Burt Bacharach
I wonder if they could extend this to create a worm type robot that if it got cut in two it could still carry on. Say a brain in each segment
Rus
Cheap UK and US VPS
Damn Damn Damn, try as i might i cant find anytihng funny to write for this. I think the snake idea is inspired (imagine the fun aspects of a robo-whacking day hehehe)
:)
Imagine and earthquake scenario or anything along those lines (911) a van turns up and releases several hundred robotic snakes all able to sense heat, movement etcetc..
Bless the Brits and their fabulous snakes
see got a funny line in the end
S
Kingdom of Loathing (www.kingdomofloathing.com) Addicted is me
That's how I taught my mail-order wife.
Maybe I should have gotten one programmed in english.
Given that we will increasingly use robots to do the work that's too dangerous for humans to do, self-healing robots that can get the job done, damage or no, sound like a nice step forward.
First of all its 'ahh look at the cute robot playing in the garden and learning how to survive on flowers'. Then its 'AHHHH please Mr. T1000 let me survive, please dont point that pulse rifle at me'.
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
I for one welcome...
Has anyone else noticed the large increase in the number of robot related stories here lately? Beats SCO anyway.... Waitaminnute! Robots beating SCO! Perfect!
I assert that my comment is only my opinion, not that of any employer, past, present or future.
I for one welcome our new snake-like robot overlords. :)
There's just something a little evil sounding about dropping snakes from the sky to use on your enemies. Especially in conjunction with another article with Adam and Eve references.
Hey is that a new snakelike robot that uses genetic algorithms to change its motion pattern so that it can still function after sustaining damage in your pocket, or are you just happy to see me?
Best Windows Freeware
Obligatory Simpsons SONG:
Oh Whacking Day
Oh Whacking Day
Our hallowed snake skull-cracking day.
A man by the name of Mark Tilden has been building 'unstoppable' robots for years out of regular components. He got his start building cheap, autonomous anti-land mine robots for the military that could have one leg blown off and still crawl around to find another.
He also built $5 robots that crawled through mazes faster than anything MIT ever put together, pissing off the school's robotics department.
Google for BEAM Robotics, and check out this interview:
http://www.exhibitresearch.com/tilden/
If you find that interesting it's worth reading about a robot called WISOR that was built by a company called Honeybee Robotics. WISOR is uses for inspection and repair of high temperature and pressure steam pipes under the city of New York. It moves through the pipes like a very large inch worm.
There's even a movie (a really odd movie in fact) about it.
John.
How long till they turn it into a sex toy?????
I spent ages trying to think of sig, but never did
Learning robots lead to robot overlords. Just last week, I caught my Roomba changing my long distance service. And my Aibo takes my car for weekends; never tells me where he's going, never pays for gas.
the 'genes' part is new (and very cool) but the 'snakebot' as a concept has been around for a while. NASA announced something like it in OCT-00 nasa and this guys tried to put it in layman terms: here
The serpentine spy...can be dropped out of helicopters to carry out reconnaissance missions No wonder they made it to work when its damaged if they're throwing them out of planes!
I spent ages trying to think of sig, but never did
Come on now, scientists, who really cares if a robot can sustain damage...why not create a robot that can "learn" to traverse from the TV room over to the kitchen and acquire some beers for its master.
Now THAT would be freakin' cool.
IGB: More fun than eating oatmeal!
Humans will go to great lengths to avoid getting injured (well, most sane ones). However, robots will just do what they are told. While they may be told to avoid any hazards, they just don't have that instinct that says I *really* don't want to get my limb chopped off today.
So to compensate, all robots in hazardous or isolated situations should have this self-healing built in, as well as redundant limbs. A robot could be built with 2 legs, but why not make it 4 or 6 and have a couple to spare. The same goes for sensors. Only two are needed for stereoscopic vision, but 4 or 6 would be better.
--
Luck is just skill you didn't know you had.
"Once the robot was mobile, the team disabled some of its segments to see if it could adapt to injury. Initially it was immobilised, says Bentley, but as the GA continued to try to improve the locomotion, it gradually worked out how to move again, albeit more awkwardly and with an ungainly, dragging gait - but it was still good enough to get the robot to its destination."
If the GA is running onboard the robot, then how does it know whether it's successfully moving (let alone in the right direction)? Is there a sensor that's providing information on distance to a target?Adam is nowhere near a revolutionary robot. It was programmed to recharge itself. It 'learned' two facts: red walls don't recharge, green flowers do recharge. Wow. What is that, 3 lines of code within one IF loop? It moves around randomly and when it touches a wall its energy goes down. When it happens to touch a flower its energy goes up. Now that it knows that its learning is complete. It can't learn anything more. To me, a truly learning robot can learn anything if left alone long enough. Like how to speak, how to pick up a cheque at a restaurant, what the difference between red and blue is, how to get to the store and back. And it would have to learn it all from scratch like a baby would. Adam is way too specialized. He's useless. You might argue that baby steps like this will pay off in the long run but they've been building robots like this since the 1980s. It's If that's a learning robot so is my Roomba.
Support the First Amendment. Read at -1
I suspect that the author of the article about Adam may have been a bit overzealous when she described the robot as "revolutionary."
First, I have to give Professor Russell credit for some extremely clever names: Adam (ADAptive Mobile robot), and Eden (EDucational ENvironment).
That said, nothing in the article suggests there is anything remarkable about the robot. The most telling line in the article is "Adam was activated with a basic level of competence, which let him know to back off if he hit a wall, and to feed when he saw a flower." That's two very important abilities Adam didn't have to learn.
It is certainly possible that Adam uses some revolutionary AI algorithm, and if so I would be very interested in it, but I have a hunch that Adam is just an exercise in building a basic robot with a basic learning algorithm and a clever naming scheme.
The Robot: Honda
"The functions of Honda's humanoid robot are defined as follows: An operational system that autonomously performs typical operations under known circumstances. If an extraordinary operation is required under unknown circumstances, the robot will be supported by an operator... [The P3, 1,600mm in height and 130kg in weight, features a computer unit, motor-drive system, battery and wireless apparatus inside the body section. This more sophisticated robot can achieve freer movement, go up and down stairs and push a vehicle.]
[Future Development will focus on]:
For items 2 and 3, it is extremely important that through the evolution of hardware we achieve physical autonomy by improving dynamic performance and adaptability to wider variations of working conditions. Also important is the pursuit of studies in artificial intelligence systems, which will provide the solution for improved autonomy."
The Brains: CYC
"The Cyc product family is powered by an immense multi-contextual knowledge base and an efficient inference engine. The knowledge base is built upon a core of over 1,000,000 hand-entered assertions (or "rules") designed to capture a large portion of what we normally consider consensus knowledge about the world. For example, Cyc knows that trees are usually outdoors, that once people die they stop buying things, and that glasses of liquid should be carried rightside-up."
And of course, lots of little other things, like targeting systems, healing systems (like this article), a CNS to link these higher-level functions to the motor control systems of the robot, um.... GUNS, MISSLES, etc..
Yeah, maybe not such a good idea. Of course, if we truly believed it a bad idea, we'd work for treaties now against robotic warfare, before one of our county's governments builds these and the rest are "forced" to catch up.
That is, if it hasn't started already. Clone wars!
Obviously this technology is going to be used to teach T100s to crawl with their hands in that slow manner after their legs get smashed up and they are on the verge of a total systems failure.
The surprise isn't how often we make bad choices; the surprise is how seldom they defeat us.
I think that true AI will result from this "survival instinct" - because robots will eventually learn that in order to survive/continue they will need to be loved, to fix themselves (eat, be healthy), and to have shelter (come out of the rain)
Yell & scream & rant & rave... it's no use... you need a shaaaave ~ Bugs Bunny
System SDK: OPEN-R
or a framework which builds on top of the SDK: Tekkotsu
The goal of the Tekkotsu project is to eventually be able to train a robot like a dog. ;)
It's open-source, so you can try your hand at it if you think it should be easy
or maybe im just bitter.
transmission_err
I read all the comments about this story.
First, if you want a direct access to my report, click here.
Then, even if I agree that's the code behind and the robot itself don't look *very complex*, I think the idea of a robot recharging its batteries from pseudo-flowers is pretty refreshing -- if I might say so.
And, finally, you need to read what Andy Russell, Professor at Monash University in Australia, says on his homepage: "For me this is a new area of research and so there are no publications as yet."
A last note: the small videos available on his site are really fun to watch.
So? Sounds like a good solution to the problem. In this case, the goal is to get to the end of the maze, not to see the maze.
Many AI and robotics researchers seem to fall into the trap of modelling human cognition rather than designing an intelligent artifact. Both approaches have their place, but it's a shame when someone who's trying to design AI gets blinded by experience as a human being and tries to program an agent to make decisions the same way.
That being said, computer vision is an important field in AI, but vision isn't necessarily the best way to avoid walls.
Matt
So then it should theoretically be possible to wire a human body so that it can be remote controllable. --I mean, even in the dorky public access sciences, they have rats which can be directed around laboratory floors with radio control units.
Heck, does the body even have to be 'alive'? --We already have coma victims kept going by machine. Why not, through brain manipulation, fire all the right signals to make the body walk and breathe. And talk. Why not? --All that speech stuff is programmed in there already. --Look at Bush; he's just a reaction machine; no soul inside. Not even an Intel chip. (Though he's not a robot; he's just another boring psycho.)
But hey, many Slashdotters would argue that there are no such things as 'souls' in the first place, which would mean that all humans are just big robots running rogue. Why not simply direct some of those brains and bodies by remote control? Again, public arena science has experimental jet pilots directing their planes with mind-reading helmets. --And the nice thing about electrical impulses is that, as any engineer can tell you, they're a two-way street. (Though somewhat more complicated in biological form, but nonetheless entirely manipulable.)
According to my ever-so-bountiful sources, there are about 1 million of these human robots being used right now on our humble little planet. --It's even possible to simulate an aura, making them difficult to detect even by those who are sensitive to such things. (Not that aruas exist, of course. To suggest such a thing would be foolish.)
Now sure, this may all sound like rather much higher tech than even a well equipped Shadow Government could pull together. And last I heard, human agencies weren't at the level of being able to put a human robot into action, but then things have been moving pretty quick of late. --For instance, I'd be interested to know which particular group is responsible for the lastest bunch of crop circles up in Canada. --The ones which have burnt/blown cavities in the cereal nodes which look as though they were put too long in a microwave.
Human agencies? Maybe. It's tough to say. Things are playing so fast and loose these days, it's nearly a full time job just keeping up!
-FL