Author Topic: Brownian motion and rotational orientation  (Read 3752 times)

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Brownian motion and rotational orientation
« on: December 08, 2006, 02:18:56 PM »
Should Brownian motion impact rotational orientation?
Many beers....

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Brownian motion and rotational orientation
« Reply #1 on: December 08, 2006, 03:24:07 PM »
No, because brownian motion can be seen as fluidic forces acting on the bot, and it's uncommon for fluidic forces to be unequal enough in the bots' small frame of reference (small enough that it's subject to brownian motion) to cause rotations.

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Brownian motion and rotational orientation
« Reply #2 on: December 08, 2006, 04:00:07 PM »
What if we made it << than the result of brownian motion on planer movement?  Fluid movement is somewhat tubulent even if you assume a small frame of reference.  In fact, all the more so if you do as thermal effects become more pronouced.  

The reason I mention this is that I think there should be something other than tie actions and voluntary rotation that can change a bot's orientation.  I have zerobots that evolve to shoot and maintain that orientation for unrelasiticly long periods.   Even if the effect is small, I'd like to see some randomness here.  I also think we should add rotational elements to collisions.
Many beers....

Offline Jez

  • Bot Overlord
  • ****
  • Posts: 788
    • View Profile
Brownian motion and rotational orientation
« Reply #3 on: December 08, 2006, 07:31:57 PM »
Is it analagous to conditions that exist for v. little things living in a watery enviroment? Would it be equally applicable to the other enviroments DB has or would you only be applying it to fluid based ones?

If it has a biological equivalent then yes, happy to have it as a new force of randomness in sim. I don't remember microbes under microscope suffering same effect, nor brownian motion come to that..
If you try and take a cat apart to see how it works, the first thing you have in your hands is a non-working cat.
Douglas Adams

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Brownian motion and rotational orientation
« Reply #4 on: December 08, 2006, 09:49:31 PM »
It would only be in effect if the user choose to have brownian motion in the sim and then the magnitude would be scaled in the same way and amount brownian motion is scaled today.  The only thing you would notice when you turned it on is that bots wouldn't stay oriented in exactly the same direction for very long, the way they do today.  Think of it as random turbulance in a tide pool or petri dish or a dog's tougue or whatever your mental model of the big blue screen is.  Either you want random mixing forces in which case I would argue you also want random turning forces, properly scaled or you don't.  If you don't you just turn it off, same as you would today...
Many beers....

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Brownian motion and rotational orientation
« Reply #5 on: December 08, 2006, 11:09:06 PM »
You're welcome to add it, but in real fluids, especially viscous ones, there are strong drag forces that are going to dampen any motion relative to the fluid.  The only way you could spin continously in a fluid is if the fluid is spinning continously areound you, with the center of the spin being you.  Not very likely.

But feel free to add an option for it, as you say there might be desire to do it.

Offline EricL

  • Administrator
  • Bot God
  • *****
  • Posts: 2266
    • View Profile
Brownian motion and rotational orientation
« Reply #6 on: December 14, 2006, 01:31:31 PM »
Wanted to raise this subject again.  What I am thinking is just adding a very (very!) small amount of random turning force in a random direction each cycle.  The strength of the force would be proportional to the Brownian motion slider value and would just be part of browniam motion I.e. bots would forever stay aligned in the same direction in a sim with no brownian motion unless they initiated their own turns or had torque applied via a tie.

There is already code in place which reduces angular momentum over time as a function of surface friction or fluid resistance.  Thus, the greater the friction or fluid viscosity, the more damping force is applied to turning momentum, be it acquired via tie torque, brownian motion or (in the future) off-center collisions with bots or other objects.  I would scale the brownian turning force such that just about any level of friction or fluid viscosity would prevent continual spinning.  Orientantions would change slowly over time, but bots would not freewheel due to my adding brownian turning forces.

As I write this, it is raining here in Seattle (big surprise).  Raindrops are splashing down into the bird bath outside my window.  Pine needles and leaves and thoise little maple seed pods have been blown into the water and the random splashing and drop action moves them about randomly.  Even as I watch, their orientation is changing, spinning slowly first one way then another.  All I ask is my evo bots be afforded the same access to realistic mixing forces as those pine needles floating outside my window...  

Please chime in if you have strong opinions.
Many beers....

Offline Jez

  • Bot Overlord
  • ****
  • Posts: 788
    • View Profile
Brownian motion and rotational orientation
« Reply #7 on: December 14, 2006, 02:09:56 PM »
Happy to have, if it includes an on/off switch to satisfy those that aren't so happy about it then all the better.
If you try and take a cat apart to see how it works, the first thing you have in your hands is a non-working cat.
Douglas Adams

Offline Light

  • Bot Destroyer
  • ***
  • Posts: 245
    • View Profile
Brownian motion and rotational orientation
« Reply #8 on: December 14, 2006, 03:25:38 PM »
I agree with Jez