Numerical Integration for Low Slip Ratios, need help

[Ending Credits]

I'm currently working on writing my own basic car physics simulation and I'm having problems with slip ratio at low speeds (the standard jittering effects). I was wondering how Racer calculates low velocity slip ratios and what other techniques there are to get rid of problems with using standard Euleurian integration.

 

[Chronus]

I found (years ago) threads with the participation of Ruud, Veble and a few others in which this issue was discussed with some detail. Try look for it, it highlights the hurdles devs and physics people have had.

In regards to euleurian integration:
try and research semi-implicit Euler integration and gauss-Seidel based constraint solvers. Good physics libs use them with very good results. You can always try explicit integration and deal with its problems (possibly come up with interesting ways of overcoming them), but I'd stick with semi-implicit methods.

Best of luck.

 

[Ending Credits]

Thanks, I'll look into it. Do know know if these threads will be on there forums (I sort of gather there has been a previous forum for Racer)?

I think for now I'll probably try using some other tyre model for lower speeds and then look at optimising for the pacejka model at lower speeds later (will need to read up more of the actual physics of tyres).

 

[Chronus]

Not here in RD, no. These are quite old threads on simulation physics forums, around 9 or 10 years ago (or even before that). The good thing about those threads is that several known developers were starting back then, so information was not abundant nor experience. Different devs/programmers trying out different things and helping each other. A lot of things can be learnt from that.

In regards to tire physics there is a ton of information from the likes of Veble, Pacejka (at least 2 books), the Millikens, Heisler, Rajamani, Fiala, Karnopp, Blundel&Harty.

Tire modelling is not about a single view/model that answers all, but trying to get as close to real life as possible - given the constraints (memory, computing power, etc).