Rodent Controll

[alan2here]

A device could be fitted to a rodents back. The device could hold a signal in front of the rodent so the rodent could see it (like a head up display). The signal would signal (for example light different colors) for different instructions such as.

 

Look more left

Look more right

Look straight ahead

go faster where you are looking

slow down

Pick up item near you in mouth

jump

 

Correctly following the instructions results in some food.

 

At first just one instruction would be given, then gradually adding more, then getting fed at the end of a successful sequence instead of after each one.

 

The device would be remote controlled and also have a small camera attached to it pointing forward of the rodent and transmitting it's signal back to a screen near the controller.

 

At this point you have a remote control rodent you can... (I will leave you lot to fill in the rest of this line).

 

Feel free to use this idea but give me credit for the concept.

I am Alan Tennant, "alan2here@gmail.com"

 

edit: saw shark topic here, maybe general chat would have been better.

[ecoli]

seems like it would be difficult to train the rodent to respond to the visual signals, and not much use once you got him to do it.

 

Perhaps you could use it to chew wires, etc?

 

I think the military (for example) might rather use AI or robots to do that kind of work.

[alan2here]

seems like it would be difficult to train the rodent to respond to the visual signals

 

Have I made a bad choice of animal? I wanted something small. (not big like a dog for example). Or a bad choice of type of signal, could be a tactile signal for example.

 

http://video.google.co.uk/videosearch?q=rodent+training

 

The first round of testing could be incredibly simple such as occasional intermittent look left signal for look more to the left, looking more to the left after it comes on regardless of how long I wait for it to happen = food. looking more to the right results in no food, (rodent will eventually move his head one way of the other).

 

 

and not much use once you got him to do it.

 

Preforming rodent show

Remote control Rodent races

Theft

Practical Jokes

 

 

I think the military (for example) might rather use AI or robots to do that kind of work.

 

Rodent:

 

is very fast

is very agile

is very versatile

finds it's own power

squeezes into spaces

bends in the middle

has good eyesight

falls great distances well

jumps

cuts

climbs

 

however they shouldn't get hurt whereas robotic vechals are more disposable.

[a Clown]

It's like a developer console.

~

/rodent_left

/rodent_kill

[antimatter]

Uh...that's suspiciously similar to the Half-Life 2 console...

[Donut.Hole]

I think that for organic subjects, cats would be best. They're very fast, know how to look out for themselves, and are much more intellligent than rodents when it comes to tactical thinking.

 

But if they were inorganic subjects, the US Military is developing robotic insects and lobsters. Their multi-legged build allows for getting into tight spots, and insects are very hardy.

[foodchain]

I think that for organic subjects, cats would be best. They're very fast, know how to look out for themselves, and are much more intellligent than rodents when it comes to tactical thinking.

 

But if they were inorganic subjects, the US Military is developing robotic insects and lobsters. Their multi-legged build allows for getting into tight spots, and insects are very hardy.

 

Tactical is one of the three typically along with strategical and operational so don’t use it to much out of context, also a lot of what could define that is situation dependent. The rodent example was to use a chip to aid in decision making based on stimulus though I don’t know how far its gotten.

[Donut.Hole]

Tactical is one of the three typically along with strategical and operational so don’t use it to much out of context, also a lot of what could define that is situation dependent. The rodent example was to use a chip to aid in decision making based on stimulus though I don’t know how far its gotten.

 

A rodent may be smarter with a chip, but raw brain power is important. The chip adapted to a more advanced brain could lead to better results. It's better to have a subject that could analyze the hundreds, maybe thousands of variables for themselves.

 

And a cat's brain can function better independently should the chip have any problems.

[alan2here]

It's harder to accurately offer a stroke and easier to offer a set size of food. From my experance cats are fussy eaters.

Edited June 5, 2008 by alan2here

[DrP]

Have I made a bad choice of animal? I wanted something small.

 

QUOTE]

 

 

No lie, during WWII, the defense research lot tried to train pigeons to guide missiles into hitting ships! Radar guided ones could get confused by flack and stuff, but the birds could be trained to look for ships. They trained them to peck at images of ships and rewarded with seed. The idea was that if they were mounted at the front of the missile, in a sort of cock pit, they could look through a perspex screen and see the target ship. The pigeon would then peck like mad at the ship through the perspex. The whole thing was set up so that the missile was steered by the pecking on the sheet. (If the ship was to the left - that's where the bird would peck and so that's the direction the missile would go - when in front the bird pecked straight and the missile went straight on etc.. Obviously the sheet would have been on some sort of movable set up sensitive to the pecking.

 

I think IIRC that they had some quite suprisingly positive results, but never actually put the idea into practice with live missiles. (you obviously can't get them to differentiate between freindly and enemy ships). Amazing!