This dynamic model shows that even when birds randomly turn individually, as if searching for their own interests, the tendency to mirror other birds results in collective turns. Thus even limited individual information can result in coordinated action, when individual agents also have the tendency to mirror.
Hit Setup. Hit Go.
You can set the number of birds before hitting setup.
If you wish, use the sliders to change the maximum angles of turns
powered by NetLogo
Individual birds mirroring the direction of nearby birds, results in flocks that turn collectively.
This dynamic model shows that enven when birds randomly turn individually, as if searching for their own interests, the tendency to mirror other birds results in collective turns. Thus even limited individual information can result in coordinated action, when individual agents also have the tendency to mirror.
Individual birds in this model follow three simple rules:
1. Mirror the turning of other nearby birds. Birds have a limited range of vision, and so they can only mirror birds within that limited range of vision.
2. Turn at random. That random turn may represent birds pursuing their own goals, e.g perhaps they see food. Individual birds turn at random, as individual birds. Yet because birds also mirror other nearby birds, flocks of birds start to form and seem to turn collectively. After each random turn, a bird flies straight for a certain distance.
3. Avoid getting to close, to avoid collision, by turning away.
The first rule, mirroring, is enough to produce flocking. Yet the addition of the individual random turns results in collective turning.
The first two simple rules are enough to eventually result in flocking behavior, if the various degrees of turn and chance of random turns are not too high. The default settings should result in flocking behavior. The avoiding of collision seems a realistic addition, and results in a more orderly looking flocking.
Note: Recently, researchers have demonstrated through simulations, that birds can "learn" or "evolve" flocking behavior as a response to predators.
Hit Setup. Hit Go.
You can set the number of birds before hitting setup.
If you wish, use the sliders to change the maximum angles of turns for:
1) turns to mirror nearby birds.
2) turns to avoid other birds.
3) turns made by birds when they turn at random.
You can change the range of vision--birds can only mirror birds they can see--those within their range of vision.
Birds form flocks and turn collectively, because birds mirror the random turns of nearby other birds. So although individual birds randomly turn (unless you set chance of random turn at 0% or very low), turns eventually are collective because of the mirroring of nearby birds.
Note that no bird has knowledge of all other birds, because they can only see a certain distance (a certain number of "patches"). Yet even this "local" vision, and local mirroring is enough to form collective turning, and grouping.
So long as the individual turns are not too random and wide, birds will eventually form flocks, and will turn collectively.
Alter the sliders to see what setting produce the most collective turning and grouping.
Chris Keane designed the mirroring and random turn procedures. Dr. Keane based the above setup procedure and the "align" procedure (including subroutines and sliders associated with "align") and the method for smoothing animation on a netlogo program created through CONNECTED MATHEMATICS: MAKING SENSE OF COMPLEX PHENOMENA THROUGH BUILDING OBJECT-BASED PARALLEL MODELS (OBPML), supported by National Science Foundation (Applications of Advanced Technologies Program) -- grant numbers RED #9552950 and REC #9632612. Copyright 1998 by Uri Wilensky. That flocking model was inspired by the Boids simulation invented by Craig Reynolds. Dr. Keane's model is very different in that a) it includes random individual turns, and b) shows that simple mirroring of other bird's direction is enough to produce flocking, c) shows flocks that continuously turn rather than all ending up going in a straight direction.
To refer to this model in academic publications, please use: Christopher Keane. Flocking as Mirrored Search (2009) and Wilensky, U. (1998). NetLogo Flocking model. All rights reserved. http://ccl.northwestern.edu/netlogo/models/Flocking. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.