Darwinbots Forum
Bots and Simulations => Bot Tavern => Topic started by: ollj on December 22, 2005, 02:57:40 PM
-
Is is possible to make a MB that grows to a Morula stadium, cells in a heap tied together without being a mess of ties?
Is it possible to let that one develop to a Blastula stadium, cells arranged in a circle, hollow on the inside?
Is it possible to let that one develop to a Gastrula stadium, the circle of cells flips into hollow-you-shape forming some kind of mouth?
Is it possible to let that one develope to a Neurula stadium, the hollow-you closes on its open part into an "O" that contains another "o" (or hollow "I")?
would it make sense at the simulators point of state?
A Blastula MB could change its shape, make itself bigger or smaller.
A Gastrula MB could swallow other bots (A Blastula could, too, by opening/widing one tie)
A Neurula MB has the ability to develop organs, bones and a central nerve system.
-
Is is possible to make a MB that grows to a Morula stadium, cells in a heap tied together without being a mess of ties?
At the moment bots connect to each other only through ties. Whether it's going to be a mess or not depends on how you program them.
Is it possible to let that one develop to a Blastula stadium, cells arranged in a circle, hollow on the inside?
Sure, but I imagine it would require a heck of a programming effort.
Is it possible to let that one develop to a Gastrula stadium, the circle of cells flips into hollow-you-shape forming some kind of mouth?
Same applies here.
Is it possible to let that one develope to a Neurula stadium, the hollow-you closes on its open part into an "O" that contains another "o" (or hollow "I")?
And here.
would it make sense at the simulators point of state?
Not sure what you mean by that.
Bot's DNA langauage allows a lot of very complicated behaviours, but it's not very easy to program. If you have some programming skills already (like many people on this board) then you can pick up the language quickly. Still, coding complicated behaviour is ..... complicated :)
-
While you could program such things, I don't see the advantage to bots to do so.
What exactly are you trying to simulate? If you can figure out what's advantagous about such things in real life, we can see if similar advantages can be derived in the simulation.
-
These are first stages of formation of higher level multicellulars. The main goal is to separate tough external surface from soft and absorbing internal surface.
These things will come in play if we ever get to the level of absorption/osmosis.
-
Is it possible to let that one develop to a Blastula stadium, cells arranged in a circle, hollow on the inside?
Yup sure is. Check out my Devincio Hexagonis (http://forum.darwinbots.com/index.php?showtopic=103&view=findpost&p=123113)
Even he is not always successful though. One of these days I will make a more reliable version. The big problem for this kind of MB is motility. It is absolute hell trying to make them move.
-
It's best to make the shapes at birth, since that's really the only time a bot can be sure of all the conditions. Shapes that are based on fractals seem to work best. You'll also need to clear the area of stray bots, to avoid accidental ties, perhaps by firing at their up command.
There is still kind of a use issue for the whole thing. A bot eats by shooting or tieing to it's prey. Any group of bots that obstruct the mouth/eye is typically less fit than the other bots. Mostly plants make the best MB's since they have no need to hunt.