Darwinbots Forum
Bots and Simulations => DNA - General => Topic started by: asterixx on June 28, 2008, 06:12:30 PM
-
I was wondering if anyone could help me improve the behaviour of this bot. I would particularly like them to be able to survive better, but also it would be neat if there was a way to make them stay in tighter clusters. Here's the DNA:
________________________________
cond
*.eye5 0 >
*623 *.myeye !=
start
*.refveldx .dx store
*.refvelup 30 add .up store
stop
cond
*.eye5 50 >
*.refeye *.myeye !=
start
-1 .shoot store
*.refvelup .up store
stop
cond
*.eye5 0 =
*.refeye *.myeye =
or
start
314 rnd .aimright store
stop
cond
*.nrg 5000 >
start
10 .repro store
stop
________________________________
Thanks.
-
The swarming behavior is caused by that first gene. It's conspec broke, so the bots now accelerate towards kin. The third gene will cause them to spin away from kin at the same time, so the net result is a poorman's swarming behavior.
There are a couple of swarming bots in the bestiary. How true to the original bot do you want to stay? The simplest tweaks would be to that 30 constant in the first gene. Set it to different values and observe the differences in behavior.
-
Well, whatever advice you might have would be helpful. One of the most important things I want to fix is that when the swarm hits a wall, it gets a little broken up. So I'm going to start with that.
-
Swarm (http://www.darwinbots.com/Forum/index.php?showtopic=1156) and Fish School (http://www.darwinbots.com/Forum/index.php?showtopic=305) are the two swarming bots I'm aware of. Fish School is based on animal minimalis, so it might be the better place to start.