But many robots communicating with each other using infrared sensors and interacting with their environment can form a group that is capable of establishing swarm intelligence to generate more complex behavior.
This seems incredibly pie-in-the-sky, no? Like someone saying "oh, we'll just hook up a trillion processors in parallel and BAMMO a human brain will result".
@Pope John Peeps II: Well, they aren't saying it'll become Skynet. The idea is that since they each have rudimentary sensor arrays and onboard processing, that by communicating and sharing aggragate data, they can form more complex analyses.
One robot can only see in its own narrow field of vision, but 500 of them can not only "see" in a full 360 view, but also have full situational awareness of themselves and other potential obstacles, and all provide each other with better pathing and decision algorithms.
It's a neural-network teamwork system. The more you add, the better the pathing and decision-making branching has the potential to become. But, as far as learning, understanding and comprehending the world is concerned... still a way to go on that score.
software simulations of swarm intelligence are simple, while the turing test remains unpassed. it's nothing more than intercommunication coordinated towards a common goal that can only be achieved (or is more easily achievable) with many separate instances working cooperatively. if all the little guys were physically tied to each other to facilitate that communication, the effect would be the same (though the wires would cause extra limitations on the possible arrangements of the bugs) but likely wouldn't be labeled as 'swarm intelligence' because the individual instances would not be physically autonomous, and would not resemble a 'swarm' but rather a single, multi-appendage device.
@nutbastard: Well true, it would let them do different things, but is that the same thing as having a more complex behaviour? They'd be able to do the same kind of tasks, just more efficiently or powerfully. But I can't understand how the simple act of hooking lots of tiny processors together would create some sort of "gestalt" ability. Or maybe that's not what the researcher is saying.
@Effing_Jedi_Master: I'm so glad that each and every post about robots gets a comment reply about Terminators. It makes me so happy that there's a wide future of humankind ahead of me which is so consistent. SO HAPPY.
I can see how Pedobear would use these child dragging swarmbots to his advantage.
Imagine; "A girly sleepover, a bed... You see little robots come from within the dark confines under the bed... The swarmbots attach themselves to unsuspecting little girls, pulling them to their doom, pulling them to pedobear."
Ok, so without knowing what is going on in the back end. The concept of it dragging a body across the floor is retarded, it's like a bad horror movie. Now my assumption is the robots really were working on their own and just going and linking up with the other robots by themselves. If that is the case that is pretty cool, but why the heck have like 20 robots doing what one or two robots could do? Guess they have money to blow. And only the swedish would come up with bright shiny colors for there robots.
@dimsum4u: It's obviously so the little robots can fit into tighter spaces to drag the child out to where the 20 foot chainsaw-fisted master robot waits.
Poor little girl, she keeps glancing at someone off camera to reassure her. I hope they paid her enough for all the therapy she will need later in life.
08/28/09
08/28/09
This seems incredibly pie-in-the-sky, no? Like someone saying "oh, we'll just hook up a trillion processors in parallel and BAMMO a human brain will result".
08/28/09
One robot can only see in its own narrow field of vision, but 500 of them can not only "see" in a full 360 view, but also have full situational awareness of themselves and other potential obstacles, and all provide each other with better pathing and decision algorithms.
It's a neural-network teamwork system. The more you add, the better the pathing and decision-making branching has the potential to become. But, as far as learning, understanding and comprehending the world is concerned... still a way to go on that score.
08/28/09
software simulations of swarm intelligence are simple, while the turing test remains unpassed. it's nothing more than intercommunication coordinated towards a common goal that can only be achieved (or is more easily achievable) with many separate instances working cooperatively. if all the little guys were physically tied to each other to facilitate that communication, the effect would be the same (though the wires would cause extra limitations on the possible arrangements of the bugs) but likely wouldn't be labeled as 'swarm intelligence' because the individual instances would not be physically autonomous, and would not resemble a 'swarm' but rather a single, multi-appendage device.
08/28/09
08/28/09
08/28/09
08/28/09
08/28/09
08/28/09
12/18/08
12/18/08
12/18/08
Imagine; "A girly sleepover, a bed... You see little robots come from within the dark confines under the bed... The swarmbots attach themselves to unsuspecting little girls, pulling them to their doom, pulling them to pedobear."
Hehe [pedobear.org]
12/18/08
12/18/08
12/18/08
12/18/08
12/18/08
12/18/08
Are there any countries where you can sedate a child to have them swarmed by robots legally?
Brilliant!!
12/18/08