Archived

This topic is now archived and is closed to further replies.

Group Transport by Ants and Robots

This topic is 5651 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

hi, I work on an ant simulation. Here is the adress of a very interesting article about the group transport by ants and robots: http://citeseer.nj.nec.com/cache/papers/cs/5520/http:zSzzSzwww.santafe.eduzSzsfizSzpublicationszSzWorking-PaperszSz99-01-008.pdf/kube98cooperative.pdf you have a sum up here : http://jmvidal.cse.sc.edu/talks/ants/allslides.html From this article I define the folowing steps(my goal is to create a Finite State Machine) 1-an ant is wandering 2-It finds a prey(How does it know that it is a prey?) 3-It positions(act to put beside the prey to grab it) 4-It tries to move the prey if successful it carries the prey until the nest if unsuccessful --> REALIGNMENT = The force applied to the prey is modified. If after several attempts the ant can''t move the prey then RECRUITMENT in the vicinity. Then go to the step 3 If unsucessful after several positionment It recruits ants in a nest(LRR) OR It gives up My problem is How does an ant decide to give up the prey? If you have any comments, solution don''t hesitate!

Share this post


Link to post
Share on other sites
Well I don''t know if your "pray objects" has different food/energy-values. For example a big worm could be worth more in terms of food/energy to the ant than a little one.

If the ant is unsuccesful at grabbing the pray it may decide to get help, if the pray has a suffiently large foodvalue. But if the pray is too small the ant may decide to give up on it altogether, that the energy required to fetch reinforcements to get the pray is too large.

I don''t know what you should use as a threshold value for fetching reinforcements though, if you use some genetic algorithm it could evolve that threshold value. Else it would be trial and error left to try for the programmer.

Good luck! (I''ll probably make my own ant sim in the near future , that explains my interest.)

Share this post


Link to post
Share on other sites
Unwise owl has a good idea I suspect. Have each ant estimate the ''value'' of the food and the ''cost'' (in energy terms) of all ants needed to get food back to nest. The decision to spend more energy than the food permits may be mitigated by the time since the last food was delivered to the nest... so if the nest is hungry, the ants might be prepared to spend more energy than the new food would give, because it will at least keep them alive. Remember that the energy load is distributed over a number of ants... so the nest may choose to allow some of the ants to die off by not letting them feed, thus making the food ''spread evenly'' among the surviving ants!

Just some ideas.

Timkin

Share this post


Link to post
Share on other sites
I''m fairly sure ants don''t give up. If they can''t pick up food whole they either cut it up or eat in on the spot. They would never just give up on a resource.




Stimulate

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
hi,

that is a good point: the ants don''t give up. the ants could be broken down into several parts after a while. What do you think about that?
To answer at your questions, no the food won''t be described in term of value/energy.

bye

Share this post


Link to post
Share on other sites
well, I am a bit late to be original, but I had to say it too Ants just dont give up... I have never heard of it, nor observed it.
The ant sees food, if it''s big, it calls people, if it''s too big it tries to break it down... simple as that.
I remember watching documentaries on "hunter" ants, and they would attack little birds and eat them. Now think about the size of a bird and the size of an ant.

Of course, it would be interesting to have the ant give up. But I just dont think this is a thing that would happen. It would probably be more interesting to see if you can simulate at what point the ants decide that the food should be cut in pieces rather than dragged away.





Sancte Isidore ora pro nobis !

Share this post


Link to post
Share on other sites