[ODE] [off-topic] Modelling aerodynamics

[Thomas Harte]

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

--=_NextPart_Lycos_0003841039362593_ID
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit

I ask these questions here because a lot of clever physics people seem to read 
this list, and I can't find a more specific alt.clever.physicians or whatever.

I, like many before me, am seeing my ODE simulated vehicles flip (think handbrake) 
and roll unrealistically often. So I've done some reading, and come across 
numerous references to the way that real cars use downforce to solve this 
problem. Therefore I wish to implement aerodynamics in my simulation - which I 
know ODE will not help with - and see if that helps me at all.

Unfortunately I'm a bit dense about this sort of thing, and can't find any 
references. Naturally I want my game to execute in real time, and it is really only 
interested in cars, so the following ideas have occurred to me :

- assume that vehicle only generates downforce when travelling forwards. Assume 
that the car will really only usually travel forwards, and simply apply a downward 
force that is in some way a function of body velocity

- slight modification of above, make downward force a function of velocity and a 
very simple function of orientation (probably just a dot product between the 
direction a car would normally consider to be forwards and the direction in which it 
is travelling)

- use an approximate model of air flow. Simulate, say, 20 well spaced air particles 
colliding with object over every time step (modelled using near-enough correct 
maths), and use these to make assertions about generated downforce.

Are these practical ideas? What do other people do?

-Thomas

Free Email at http://www.lycos.co.uk


--=_NextPart_Lycos_0003841039362593_ID--