Slashdot Mirror
Learning Autonomic Robots
Daath writes "The 27th of March, Professor Noel Sharkey et al starts a colony of living robots. 15 predators and 6 prey. It's an experiment in artificial evolution out of the Creative Robotics Unit at Magna. Here's a quote: 'The Living Robots have one goal, to obtain enough energy to survive and breed. The prey find their food from light sensors within the arena, while the predators feed off prey by stalking and chasing them before sucking away their power.'
Magna has two articles, 'Predator and Prey Robots set up home at Magna' and 'Ground breaking Robotics experiment previewed'. "
When MIT's AI lab was getting started (around the 1960's I think), they got really interested in robotics. Now, this isn't obvious to me. What does intelligence have to do with robotics? Doesn't a Turing Test (which by its nature involves bits, rather than physical world) more accurately reflect the nature of intelligence? Well, the thinking at the AI lab was that robots were faced with a much more realistic picture of what humans had to navigate. That robotics by its nature involves dealing with uncertainty, with unpredictability, and so building a virtual intelligence wouldn't really illuminate the real problems of intelligence.
"...spectacular 30 minute live action show - complete with atmospheric lights, smoke and music."
"Each show will begin in darkness. Dramatic music will flood into the arena as guests prepare themselves for the spectacular light, sound and science show."
Maybe I'm just a little jaded right now, but this sounds more like a circus show instead of a serious scientific experiment. I'm sure these are very complex robots, and the underlying idea is very interesting, but the whole BattleBots spin on it seems to trivialize the work. Now of course if he signs up Carmen Electra.......
I posted to
Sigh! This is outrageous. Cognitive Science as opposed to Good Old Fashioned AI was I think one of the most sensible currents in recent day CS research. And now just to discredit whatever sensible, realistic research occurs in these fields within the academic community here is Yet Another Crazy AI project grabbing front page visibility at
Prof..you give CS a bad name !
**
It depends on the relative food source/requirements. We normally consider 'predators' to be large animals, which mean a lot of prey, but if you think about it, there are instances where the ratio is reversed. One cow can support thousands of fleas, ticks and other small beasties.
Think of it like the team who found the Titanic. Roughly zero scientific learning, but the public interest in it brought in enough money to fund development of the remote vehicles. Once the cameras point to something else, they're left with some expensive new toys to use to do some real work.
Nope, no sig
then all of the surviving robots get paired off randomly
...which makes this pretty stupid. The whole idea of evolution is built upon "selection" i.e. the robot that does best has most offspring. Just looking at survival rate is a measure for measuring fitness, but it's too crude a method for improving ones genes. Besides that now every surviving bot has the same amount of fitness (offspring). That seems to be some binary kind of selection which I at least have never come across in real life. Randomly mixing genes is therefore 'not' a good method to mimick nature.
If an experiment works, something has gone wrong.
Object oriented design is perfect for this sort of thing. I did a simple experiment in Java, where predators, prey and food pellets were objects. Each object could have many different characteristics which chould be set when each object was spawned, which kind of mimics evolution. Also, if the logic in an object needs an upgrade (ie: The preditors are not too bright) it is easier to make modifications to the program instead of rebuilding a real robot.
I guess anything with real robots has a certain coolness to it, but any serious research in AI is better done in software simulations (not that I did any serious research, I was just learning Java and OO design).
with no real scientific significance. The methodology is way off.
Most obviously, you need far more prey than predator in the "ecological" system they propose.
Secondly, there needs to be far more of both types of robots in the system, and they need to have some minimal learning abilty. For instance... take over a whole building, placing feeder stations in random locations at a level where the prey can get to them. Make the prey different in design from the predators as well, so that they can find refuge in certain locations where predators cannot go. Make the prey somwhat competitive or territorial as well, so that they all don't end up packing into proteted area's.
Both predator and prey should breed (sharing their learned AI behavior and passing it down to their progeny), with new robots being introduced upon successful completion of the breeding cycle.
Make the feeder stations for the prey have finite resources... meaning that it can only put out so much food (power) each day.
Program two entirely different base sets of "instincts". Prey animals look for food and safe area's, congregate together in packs for protection, and seek refuge area's. Give them a docile pack mentality so they form bands and don't accept outsiders (other than breed "babies").
For the predators, make them solitary and aggressive (canibalistic)to other predators except when they need to mate (arbitrary cycle of time).
Make the prey have a minor defense from predators from the front only (as a programmed behavior). When attacked from the front by predators, the predator is stunned and looses some energy and has to learn to attack from the rear.
Give both a power management scheme as their central programming that hooks into their behaviors. They learn through this as part of tehir AI since their goal is twofold to breed and survive by acquiring energy.
I am sure there is more... it just seems that the experiment is really more of a press deal than anything else. Give us real science.
I am pretty sure that building these robots with complicated behaviors (as listed above) is technically not feasable just yet on a minor budget and due to programming an size constraints. However, the results would be interesting to see.
It would be more cost efficient, but a lot of useful data might be lost. Robots that must interact with the real world have to deal with the messiness and uncertainty that it entails. e.g, a predator robot can lose track of its prey due to a faulty sensor, or an interfering signal, or its wheels might slip on the floor, thus allowing hte prey to escape. None of these would be present in the simulated world of the program you are suggesting. There's more to research with physical robots than "eye candy."
the RESULT depends on the goals you DEFINE:
'The Living Robots have one goal, to obtain enough energy
to survive and breed.'
thus, it is not like evolution at all, but comes with
a built-in BIAS that DEFINES their evolution.
"Think again before postulating the drive to self preservation
as the cardinal drive in an organic being. A living thing desires
above all to vent its strength - life as such is the
will to power - self preservation is only one of the indirect
and most frequent consequences of it". (Freidrich Nietzsche)
Uhh... why? If the purpose of the project is to demonstrate a particular characteristic of a biological ecosystem, then trying to artifically replicate as much of that ecosystem as possible is critical. If, on the other hand, you're trying to engineer a different ecosystem based on some basic rules from an existing ecosystem, then an identical reproduction doesn't matter.
The situation is similar to studying birds in order to understand flight. For a long time we assumed that the only way to fly was to try and identically replicate the flight of birds - i.e. flapping wings. It was only when we started to understand the basic components of flight - that the shape of the wing allowed the exploitation of the bernouli principle - that humans began to fly. In fact, to the best of my knowledge, there is no flapping wing aparatus that can enable human flight. In other words, the most effective way towards human engineered flight was to eliminate some of the factors in biological evolved flight.
So, even if this experiment isn't a complete biological replication, it doesn't matter. It's simply studying one aspect of biology and intelligence in order to see what things are/aren't important in being able to engineer an intelligence.
$.02
Key to financial independence: Spend less than you earn. Save and invest the difference. Do it for a long time.
No.
The you replied to a post that (unfortunately) used parasites as an example but the poster was correct. A single cow can also feed dozens of foxes (foxen? foxi?), coyotes, and gee, humans. When was the last time you sat down to dinner and ate an entire cow?
If this simulation is using more predators than prey then that is the ratio that is called for. I'm pretty sure the scientists that put this together know what they are doing since they were smart enough to build the robots in the first place.