Wheel Friction Question

I understand the basic physics of static versus dynamic friction, and lateral versus longitudinal forces, but the tire force function confuses me. Does anyone have any insight on these parameters?
  • AsymptoteSlip
  • AsymptoteValue
  • ExtremumSlip
  • ExtremumValue
  • StiffnessFactor

[314 byte] By [BenAxelrod[CoroWare]] at [2008-1-9]
# 1
Ageia gives some information about it in its documentation (Lesson 702/703):

Force
^extremum
| _*_
| ~ \ asymptote
| / \~__*__
| /
|/
> Slip

  • ~extremumSlip - extremum point of curve (default 2.0)
  • *extremumValue - extremum point of curve (default 0.01)
  • ~asymptoteSlip - point on curve at which for all x > minimumX, function equals minimumY (default 2.0)
  • *asymptoteValue - point on curve at which for all x > minimumX, function equals minimumY (default 0.01)
  • stiffnessFactor - scaling factor for tire force

Basically, it is calculated first how much the wheel is slippling and then the corresponding force it taken from the TireForce function. The extremum parameters describe this function, as shown in the graph.
Haven't experimented with the stiffnessfactor so far.

Hope that helps
Tom

Just for reference here a code example:

Code Snippet

TireForceFunctionDescription lngTFD = new TireForceFunctionDescription();
lngTFD.ExtremumSlip = 1.0f;
lngTFD.ExtremumValue = 0.02f;
lngTFD.AsymptoteSlip = 2.0f;
lngTFD.AsymptoteValue = 0.01f;
lngTFD.StiffnessFactor = 100000.0f;

WheelShapeProperties w = new WheelShapeProperties(...)
w.TireLongitudalForceFunction = lngTFD;



ThorDe at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...
# 2
Thank you, that helps a lot. Some follow up questions:

1) I have seen lots of people reference the Ageia documentation, but I do not know where to find it. Is it freely available?

2) What are the units for these parameters?

3) Am i correct in assuming that the point (extremumSlip, extremumValue) is the coordinates of the peak in your plot? And the point (AsymptoteSlip, AsymptoteValue) is the point where the right side of the peak meets the horizontal line?

Also, it would help me out a lot if you could show me an almost frictionless wheel.

BenAxelrod[CoroWare] at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...
# 3
Hi,

1) For Ageia documentation you have to apply for developer support on their homepage. After they approved your application you will have access to SDK, FAQ and documentation.

2) The documenation will tell you then that basic units are kilogram, meters and seconds. Everything else follows from these units, i.e. N = kg*m/s^2 and so on.
So Force should be measured then in N. Slip is calculated as speed difference and from this interpretation it should be m/s.

3) TireForceFunction is a hermite spline defined by the coefficients which is defined by the Extremum and Asymptote values. I am not the expert here but your interpretation of the plot (which is from the docu) is my as well (and see the link below).

Had a quick look around and found a comment on the stiffness factor in the API doc of another engine:
-higher values improve grip
-a zero value will disable forces, i.e. you will then have no friction.

I tested both cases, so in the latter case if you insert a vehicle and set all stiffness values to zero, you won't be able to move it via wheel turning. No surprise though, because there is no friction.

So if that is what you want, set the stiffness factor to zero.

Tom

ThorDe at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...
# 4
Thank you. Setting the stiffness to a very low value worked great.

I would like to note that if you specify a tire force function, you must manually set the 4 standard parameters (AsymptoteSlip, AsymptoteValue, ExtremumSlip, ExtremumValue). Stifness seems to be optional. If they are not set, _wheel.Wheel.AxleSpeed, _wheel.Wheel.MotorTorque, and _wheel.Wheel.SteerAngle will be null when the entity gets updated.

Also, a previous post said that these relations must hold:

ExtremumSlip <= AsymptoteSlip

ExtremumValue >= AsymptoteValue


I have found that this is the only relation that must be held:
ExtremumSlip < AsymptoteSlip

BenAxelrod[CoroWare] at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...
# 5

If the Force is measured in N, 0.01 or 0.02 N are very small for a force and I think that there is also a relationship with the stiffness. I have readen the Ageia documentation and the units are not done. Is there a way to know precisely the unit of the force considering the stiffness and, if the slip is realy in rad/s or in radians or degrees (like in the publication purposed in the Ageia documentation: Race Car Vehicle Dynamics by Milliken et al.) ?

Thanks.

Eric

LUCETEric at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...
# 6
Hey,
its all about putting bits and pieces together, so I did some further research and found this model quite helpful.
The Ageia forum supports the fact, that the parameters are normalized.
Stiffness factor can be seen as maximum load, i.e. stiffnessFactor= 50000 equals 50000N.

Tom

ThorDe at 2007-10-3 > top of Msdn Tech,Microsoft Robotics Studio,Microsoft Robotics - Simulation...

Microsoft Robotics Studio

Site Classified