Scaling Netlogo Model to Real World Dimensions

[Emily W]

Hi,

My lab is planning on running laboratory experiments that can be compared to a Netlogo model of ants that I built, and my colleague was wondering, "How is the scale of the world measured?  Is it in hypothetical centimeters? 

If not, do we have to do some sort of calculation of how large the actual ants are to the arena and then scale the simulation from there? (Or vice versa and put the food in relative distance to what is seen in the model)."

How are the agent sizes (especially for unique shapes, like the bug shape) compared to patch size? For instance, my model ants are set at size 2.0, so does that mean that they are set at twice the size of a patch, but in a bug-shape?

Thanks!!

[John Chen]

Yes, your assumption is correct.

Emily W <liftu...@gmail.com> 于2021年5月13日周四 下午2:35写道：

--
You received this message because you are subscribed to the Google Groups "netlogo-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to netlogo-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/netlogo-users/a11a32d6-b040-4995-855c-a432d02d9878n%40googlegroups.com.

[Emily Warsavage]

Which assumption - the one about the world dimensions, or about the agent sizes?

[Michael Tamillow]

Hey Emily,

This is a very difficult question. Models are simplifications of the world, and this is really dependent on the modeling choices you make. You could make some estimates like how long it takes for ants to travel over a certain length vs how long it takes for them to travel in your simulation. Based on my recent experience with ants they tend to speed up when threatened. (and I would imagine when busy!) Even Netlogo "ticks" aren't exactly times, so you have to pick units for both ticks and patches that gives you a pretty good estimate of how the ants are traveling to go from one place to another.

[Brad Reisfeld]

Hello Emily,

For an ant, there will be the display size and the simulation size. The graphical interface can show you the ant at whatever size you'd like (in units of patches); however, the simulation engine considers your ant to be a zero-sized point at a certain x, y coordinate in the world.

The size of the Netlogo world (the mapping between patch size and physical dimensions) is up to you.
The number of patches in the world is established by setting min-pxcor, max-pxcor, min-pycor, max-pycor.

Assume that your lab experiment occupies 1 meter by 1 meter. If you'd like each patch to represent 1 cm x 1 cm, you could set min-pxcor and min-pycor to -49 and max-pxcor and max-pycor to 50.

The same principle goes for ticks. A tick could represent any unit of time (e.g., a second, a decade). So, based on some activity of interest for your ants (e.g., crawling or building a hill) and your chosen world size, you could establish what each tick should represent in terms of a reasonable clock time.

-Brad