Nice find Juls. I'm already using another PDF by that same guy for bolting brush model combination slip on top of Pacejka pure curves, but that document explains a much broader application of brush models. Could be worth reading to try and do away with Pacejka all together at some point.

The key thing you need in a sim is a fair degree of configurability, which Pacejka's MF certainly does provide, even if not in a convenient form. Will have to see how much control you have over the curves of a complete brush model.

Split these posts into a new thread because it's a) tidier and b) people can find these particular posts more easily at a later date, rather than being buried in an ~1100 post thread

Quote from Juls :

Good comprehensive thesis:
http://www.control.lth.se/documents/2007/jsvenPDH.pdf

thanks a lot for this read

Hello,

With the UF1 (Open Differential)

When I turn it and the inside wheel has more grip, the power passes through it (what is normal)

But it gives Bad behavior in games

The power goes to the ground so it should skate and suddenly I find myself in understeer

More, on traction when the front tires have more grip, it keeps the directivity

is totally unrealistic

Open diff - both wheels get the same torque at all times. The inside wheel has LESS grip, and thus defines the maximum torque that can be applied to the ground.

open diff :


if a wheel slips, the power passes through it.

Yes, that's what I said - both get the same torque, which is limited by the wheel spinning.

ok, I just copied and pasted formula given by Racer for Fy

Quote :

Fy=Fy0*sqrt(1-(Fx/Fx0)^2)
Where:
- Fy is the resulting combined slip lateral force
- Fy0 is the lateral force as calculated using the normal Fy formula
- Fx is the longitudinal force as calculated using the normal Fx formula
- Fx0 is the MAXIMUM longitudinal force possible (calculated as D+Sv in the Pacejka Fx formula).

and combined it (F combined)
and I must say it looks quite similar to my previous one (Pacejka combined)

FYI - in the second one I separately diminished Fx as vector by Fy and zeroed when negative, and respectively done with Fy, then combined both as vector components.
In the "Pacejka combined?" I just used in vector sum (Fx-Fy)^2+(Fy-Fx)^2

Quote from AndRand :

Direction (and heading ) come from x and y components

btw. joke for physics geeks:

A man jumps on the bus and asks (bad luck a physician):
- I s this direction to railway station?
- yes it is.
- and how many bust stops?
- 3
- thank you.
After two bus stops he asks:
- so this would be the railway station bus stop?
- no, now you have 5 bus stops
- HOW COME?! YOU SAID IT WAS DIRECTION TO THE RAILWAY STATION!
- yes it is, but the heading opposite

hehe, I catch it, even though I failed every single physics exam I made in my last year in high scholl

Quote from JeffR :

In my opinion, making a significant improvement over the current tire physics mode in LFS will end up going beyond what an individual can accomplish, without access to good documentation of the internals of newer physics models, in addition to sorting out which of the newer models will work well in a racing game environment. What I don't know is if access to such models exists.

Quote from tristancliffe :

JeffR hasn't got a clue what he's talking about. He fails to realise it is simply a man-hours thing. He's also under the impression, it seems, that Scawen uses a preexisting model for his physics.

It's not a man-hours thing when the number of man-hours exceeds the lifetime of an individual. It doesn't take much to be aware of the fact that there is a significant amount of research being done in the field of tire physics modeling, and that this research has gone beyond what any single individual can do on his own without the exchange of information between others (corporations and universities) also involved in this research.

Based on the time it's taken already, it should be obvious that it's difficult. I don't know what model LFS uses, but I would assume that it's based on information from other models.

What I do have a clue about is that many things are complicated enough to go beyond what an individual can accomplish on his own without input from others. As a programmer for a couple of computer peripheral companies, I've been exposed to a few university based research centers, such as http://cmrr.ucsd.edu where professor Jack Wolf and some of his PHD students visited and exchanged information regarding signal processing used in the channel designs for tape and hard drives, and professor Weldon http://www.ee.hawaii.edu/faculty/detail.php?usr=27 who exchanged information to be used with error correction codes, an area that I specialize in.

No individual could hope to replace the knowledge gained through man-decades of research into these or similarly complex areas, and I'm sure that tire physics modeling is just as complex as many other fields of research.

Quote from JeffR :

The issue with a smaller development team is isolation from the rest of the development community and/or access to real world data. The larger teams generally have some ongoing influx of people and/or data that helps them keep up to date with current methods and to avoid any pitfalls already discovered by others in the community.

I don't recall any LFS dev saying they're totally ignoring everyone and everything else and re-discovering the virtual wheel. For all we know Scawen might have a kidnapped physicist in a box in his basement, with a Geiger counter, a bit of radioactive material, so small that perhaps in the course of one hour one of the atoms decays, but also, with equal probability, perhaps none...

Quote from JeffR :

No individual could hope to replace the knowledge gained through man-decades of research into these or similarly complex areas, and I'm sure that tire physics modeling is just as complex as many other fields of research.

That's right... but bear in mind that purely scientific modelling and then simulation has to be exact nearly to the empirical precision of survey (as I know simulations done that way have to use ie. the same data input as modelled car and give just the same results as those obtained from that car). But it doesnt have to cope with grid of 20+ cars running on every (average) PC logged in. Therefore, yes, for racing sim it is important to get the scientific results and crucial to simplify complex modelling.

edit: plu another colorful diagram - I still wonder how to interpret this

I just wanted to see what this "magic formula" by Pacejka is all about.

So I incorporated Pacejkas tyre characteristics and combined F vector once with this formula, second time with reduction of vector equation I used: modulo(Fx-Fy)*sqrt(2)

I must say... quite different

Quote from AndRand :

modulo(Fx-Fy)*sqrt(2)

Err... What is that for? Modulus of Fx - Fy * sqrt(2)? No wonder the maps are a bit goofy. If Fx and Fy are equal to or multiples of each other, then F=0. There isn't going to be a non-zero combination of slip angle and slip ratio that results in 0 force of course. See the valleys in the graphs that are at 0 force? That's probably why.

You ought to be thinking more along the lines of Pythagoras, probably. I'm curious how and why you came up with that formula though.

If you're trying to make a 3D map of this, all you need to do is map Fx according to its constants and Fy according to its. The 2D vector that you get IS the resultant by definition. No need for doing something extra on top of this. However, as Bob suggested, you then could check the length of this resultant vector and then shorten it so it is no larger than the radius of the friction circle.

Waow this is a really interesting stuff! Nice to see people taking time to do research about tyremodelling. It's really cool to read all this, thanks!

But as the friction circle is actually an ellipse, a slightly more complicated equation is needed to provide the limiting values. Otherwise the car's will have too much grip in certain common conditions (like braking and turning)

Quote from jtw62074 :

You ought to be thinking more along the lines of Pythagoras, probably. I'm curious how and why you came up with that formula though.

I assumed that the first is diminished by the second (maximum force available for one component)
sqrt((Fx-Fy)^2+(Fy-Fx)^2)...

Quote :

If Fx and Fy are equal to or multiples of each other, then F=0. There isn't going to be a non-zero combination of slip angle and slip ratio that results in 0 force of course. See the valleys in the graphs that are at 0 force? That's probably why.

And it makes sense (maybe with some scaling not to go beneath any force available at that moment) - if you have slip angle getting the highest force and you add longitunal slip you will loose grip very quickly. And with "magic formula" when you are on lateral peak force adding slip ratio to peak longitunal will result with bigger force overall

edit: I suspect there is something very subtle going on and it doesn look like "magic formula". For now I got the ellipse and zeroed everything around it (attachment)

Todd, correct if I am wrong - I was curious about these Pacejkas formulas, that's why I posted them and it looked to me that they are strictly geometrical just to fit the empirical data. Therefore coefficients are also strictly geometrical, not derived from friction theory and that's why friction-related coefficients are like: D+Sv (maximum force).
So when they dont fit the data, some "magic" is needed More complex equation to fit the data... so the thing is: to fit diagram to empirical data, right? So the empirical data for both slip ratio and slip angle changing are crucial...

Why is it incredibly difficult to turn the steering wheel in a stopped vehicle, but easy when the vehicle is moving?

I'd say because when you're standing still, you have to twist / overcome the static friction of the entire contact patch, whereas when the tyre is spinning you twist the "current" contact patch just a tiny amount before a "new" bit of rubber comes along (from the rotating tyre) that is already correctly aligned with the new steering angle. So you're kinda limiting the amount of actual rubber twisting required because the rubber you're twisting is constantly being exchanged with new pre-aligned rubber. At least that's how I envision it after pondering a few minutes. Hope it makes sense.

Quote from wheel4hummer :

Why is it incredibly difficult to turn the steering wheel in a stopped vehicle, but easy when the vehicle is moving?

I noticed the other week if the car is sitting in the garage and you move the wheel side to side there is no ffb but if you apply the brake the ffb will kick in for some reason.

That's because of the scrub radius of the suspension / wheel mount. Basically when you steer the wheels they don't pivot around their exact centre (viewed from top), but around an axis that is slightly offset towards the car. This means when you turn the steering wheel the tyres actually roll along a tiny arc. If you lock the brakes, the tyres won't be able to roll freely and you have to scrub them along this tiny arc, which in turn causes the resistance.

This is very noticeable in LFS because apparently the normal twisting/turn resistance of the rubber is not modelled, so turning the wheel without brakes has almost no resistance, which is wrong.

It's wrong, but it doesn't matter in a simulation, as the effect at racing speeds is negligable.

Quote from tristancliffe :

It's wrong, but it doesn't matter in a simulation, as the effect at racing speeds is negligable.

Can also be (partially) explained away with power steering...

Quote from AndRand :

if you have slip angle getting the highest force and you add longitunal slip you will loose grip very quickly.

no you wont
you might lose some amount of lateral force to longitudinal force but the overall combined force will if anything increase and most certainly it wont drop to 0