rec.autos.simulators

Kinetics ... question

Sébastien Tixie

Kinetics ... question

by Sébastien Tixie » Wed, 10 Oct 2001 05:15:02


> On Mon, 08 Oct 2001 19:26:46 +0200, Sbastien Tixier

> >your solution seems to simple for me or maybe i miss something else :o(

> You're talking into yourself, Sebastien? ;-)

> Never been a star in solid body maths, sorry, can't help you much
> here. Restraint forces (torques rather) are something I will get into
> when I try to couple 3 differentials in a row.

hehehe, trying to add 4WD cars in you sim ? ;o)

thanks ,good luck too ;o)

--
Sbastien TIXIER - Game Developer
Dynamics and Car Physics
Overall GPLRank -44.24
Monster GPLRank -114.44

J. Todd Wass

Kinetics ... question

by J. Todd Wass » Wed, 10 Oct 2001 06:25:12

  Wish I could help you Sebastien, but I haven't begun any collision detection
stuff yet.  Still developing a new suspension model here.  Have you read
Baraff's paper?  That covers quite a bit of stuff, although most of it is over
my head at this point.

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

SimRace

Kinetics ... question

by SimRace » Wed, 10 Oct 2001 07:06:38

Thanks for pointing that out to me Johnny, I'd never figured that out on my
own, sheesh.

Ask a simlpe question, get a complex troll.





Mike Stanle

Kinetics ... question

by Mike Stanle » Wed, 10 Oct 2001 07:33:41

Actually it looks like you're trying to do the same thing I'm currently
working on (should have spotted that sooner).

Correct me if I'm wrong, but the way I see it you have 2 rigid bodies
connected by a constraint. A force gets applied to 1 body. You want to
calculate what forces are applied at the constraint point in order to keep
the bodies together.

If so then you need to look at the Lagrange Multiplier technique, which does
exactly that. The maths is quite intensive at first, but once you get over
that you just have a bunch of formulae to plug numbers into :)

The best online resource is probably
http://www-2.cs.cmu.edu/~baraff/sigcourse/.

The main problem I've found so far is that this stuff is usually described
in terms of systems of particles rather than rigid bodies (see my other post
in the thread "Car Physics: (OT) Lagrange Multipliers").

Hope that helps,

Mike.


Sebastien Tixie

Kinetics ... question

by Sebastien Tixie » Wed, 10 Oct 2001 14:53:23

Thanks for the link !


> The best online resource is probably
> http://www-2.cs.cmu.edu/~baraff/sigcourse/.

--
Sebastien TIXIER - Game Developer
Dynamics and Car Physics
http://www.eden-studios.fr
GPLRank Normal:-44.24 Monster:-124.44
Ruud van Ga

Kinetics ... question

by Ruud van Ga » Wed, 10 Oct 2001 19:34:30

On Mon, 08 Oct 2001 22:15:02 +0200, Sbastien Tixier


>> Never been a star in solid body maths, sorry, can't help you much
>> here. Restraint forces (torques rather) are something I will get into
>> when I try to couple 3 differentials in a row.

>hehehe, trying to add 4WD cars in you sim ? ;o)

Yep. :) Seems rally people are a majority of users of Racer somehow.
And keyboard support (yugh!).

Thanks. :)
I do think I'm missing some posts here, damn. Can't follow the thread
at all.

Ruud van Gaal
Free car sim  : http://www.marketgraph.nl/gallery/racer/
Pencil art    : http://www.marketgraph.nl/gallery/

Sebastien Tixie

Kinetics ... question

by Sebastien Tixie » Wed, 10 Oct 2001 20:52:42

We were talking about constraint for rigid body, it seems that i need to
implement... err.. undestand first, the Lagrange multiplier and Jacobian
matrices ;o)... it's a jump 8 years back for me ;o)

--
Sebastien TIXIER - Game Developer
Dynamics and Car Physics
http://www.eden-studios.fr
GPLRank Normal:-44.24 Monster:-124.44

Ruud van Ga

Kinetics ... question

by Ruud van Ga » Wed, 10 Oct 2001 21:52:11

On Tue, 09 Oct 2001 13:52:42 +0200, Sebastien Tixier


>We were talking about constraint for rigid body, it seems that i need to
>implement... err.. undestand first, the Lagrange multiplier and Jacobian
>matrices ;o)... it's a jump 8 years back for me ;o)

That's a jump 8 year AHEAD of me. ;-)
The Baraff papers seem the bible from what I've learned in Internet
browsing for this kind of stuff. Also, the Erik McKenzie Lowndes
thesis uses constraint forces I think, or at least Lagrangian
mechanics.
Get that at
http://www.lib.ncsu.edu/etd/public/etd-373202299842941/etd.pdf

It's a description of a car simulation with even some AI things in it
etc.

Ruud van Gaal
Free car sim  : http://www.marketgraph.nl/gallery/racer/
Pencil art    : http://www.marketgraph.nl/gallery/

J. Todd Wass

Kinetics ... question

by J. Todd Wass » Thu, 11 Oct 2001 07:50:55

  Ruud,

  Note that on page 117 the author says he didn't use any tire damping because
it's small compared to the suspension damper effect.  Good news for you :-)

  Page 122 shows that he does NOT use a constant rate tire spring.  This makes
sense because the contact patch size shouldn't grow linearly as the tire "disc"
disappears into the ground.  The tire rate you've been using should start out
very small and increase as the tire sinks, don't you think?  Maybe that could
help vertical stability if you go back to using the tire force instead of the
suspension force at some point.

  A couple of pages later concerning low speed oscillations in tire slip:

"Bernard and Clover found that oscillations occur in the lateral and
longitudinal slip at low velocitiites.  To eliminate these oscillations it is
necessary to inclde a damping term which is only activated when the velocity is
below a certain threshold (0.15 m/s is suggested) and changes sign from one
integration time step to the next.

  Search for equation 7.3.9 for their version.  I recall you saying when
getting that SAE paper(955011?) to work, you needed a damping ratio some 27/37
times critical.  Where you only implementing it under the above conditions?

  Sorry if it's a little off topic, had to mention this.

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com

Ruud van Ga

Kinetics ... question

by Ruud van Ga » Thu, 11 Oct 2001 20:47:45


Yes. :) In fact, I read this before in some other book. Although the
*** damping is in principle way higher than a suspension's damper I
think when thinking of pushing my finger into *** and letting go,
the length at which this happens is much smaller in the case of the
tire, so probably that's why the tire damping can be neglected. I
wasn't going there anyway, afraid of explosive integrations when
damping small chunks of ***; far too high oscillation frequencies
for my poor Euler integrators. ;-)

Hm, yes, thanks for noticing this. Still, that would mean an extra
parameter again, which is something to avoid (because it's one of
those 'guess-it' parameters :) ).
Major thing for now is to get Pacejka editing into the world, because
people are still complaining about the incredible loss of grip when
tire finally lets go. Even boosting up gravity by a factor of two for
playability! ;-)
BTW a nice tip perhaps for the 'arcadish' feel that was talked about a
short time ago; next to flattening the end of the tire force curves
(keeping it high near the peak), you could try using double gravity,
next to higher body inertiae. But let's not go to NFS physics too
much. ;-)

That was 950311 for those who want to go through their files. I read
this too (even posted about it here) and mostly because Eric states
that 'the damping term changes sign each integration step'. This I
couldn't find in the original SAE950311 document. Still, the lateral
stiffness you need in that formula I can deduce from the Pacejka
constants, but for the longitudinal stiffness, it seems not really
defined. I mean, in Genta's book there's something about C, IIRC,
being the stiffness (coefficient?), but no such thing for Fx
(longitudinal). Still, I use it in the same way, as all Pacejka
formulae are basically the same in their form, and just use shaping
parameters (and in fact match physical meanings, which I previously
didn't think; not so much Magic after all).

Thanks for mentioning it. I've got a tasklist from here to Tokyo, but
things like this always stick in my mind (for the future), since the
instability ordeal at low speed was such a pain. :)

Ruud van Gaal
Free car sim  : http://www.racesimcentral.net/
Pencil art    : http://www.racesimcentral.net/

J. Todd Wass

Kinetics ... question

by J. Todd Wass » Tue, 16 Oct 2001 07:08:36

  I saw that happening at the forum.  Blasphemy!!!   I'll do no such thing  ;-)
  Well, maybe an arcade mode would be necessary....

Todd Wasson
---
Performance Simulations
Drag Racing and Top Speed Prediction
Software
http://PerformanceSimulations.Com


rec.autos.simulators is a usenet newsgroup formed in December, 1993. As this group was always unmoderated there may be some spam or off topic articles included. Some links do point back to racesimcentral.net as we could not validate the original address. Please report any pages that you believe warrant deletion from this archive (include the link in your email). RaceSimCentral.net is in no way responsible and does not endorse any of the content herein.