Performing quasi-static vehicle simulations. (Simscape/multibody)

16 Ansichten (letzte 30 Tage)
Good day everyone.
I would like to model "quasi-static" simulations in Simscape/multibody. I created a vehicle model. Works perfectly well in shaker rig type simulations but I would also like to apply lateral/longitudinal accelerations (using variable gravity/time varying option within mechanism configuration). I can't figure out how to properly constrain the wheels (to the "ground") so that the car doesn't slide away under lat/long acceleration loads. I tried using MF tire block and modify tire property file to get artificially high Mu (higher than applied acceleration) but car model is still sliding away. Interestingly enough the sum of four tire lateral forces (applying lateral acc load to the model) is close to overall mass of the model but still slightly lower i.e. not enough to constrain the car from sliding off. Trying to constrain the wheels via joints didn't work either. Other than having problems with "model not assembled", even if I manage to assemble it without errors it would crush when trying to run the simulation.
Any suggestions ideas would be highly appreciated.
Thank you.
Ted

Akzeptierte Antwort

Jack
Jack am 8 Mär. 2025

One common approach for reducing wheel sliding in quasi‑static simulations is to improve how the tire–ground interaction is modeled. The MF tire block may not be capturing enough frictional force under quasi‑static conditions, especially when using variable gravity to simulate lateral/longitudinal accelerations. You might try the following strategies:

– Instead of relying solely on increasing the friction coefficient in the MF tire block, consider using a more detailed tire–ground contact model (for example, one from the Vehicle Dynamics Blockset) that can enforce a no‑slip condition more effectively.

– Rather than modifying gravity to generate acceleration, apply external forces or motion inputs directly to the vehicle chassis. This approach can help maintain proper tire contact without inadvertently reducing the effective friction.

– Explore adding constraint blocks (such as a planar or no‑slip joint) to restrict wheel motion relative to the ground while still permitting wheel rotation. This method forces the tires to follow the ground trajectory and can help prevent sliding, though it may require careful tuning to avoid over‑constraining the model.

– Review your solver settings and simulation tolerances. For quasi‑static simulations, a slower solver with tighter tolerances may better capture the small differences in forces that are critical for maintaining wheel grip.

These adjustments should help you achieve a simulation where the tire forces are sufficient to counteract the applied lateral/longitudinal loads.

Follow me so you can message me anytime with future questions. If this helps, please accept the answer and upvote it as well.

  2 Kommentare
Tudor Miron
Tudor Miron am 8 Mär. 2025
Many thanks for your prompt reply Jack. I will try suggested strategies.
One thing I didn't mention in my above question. It would be desirable that track width and wheelbase are still free to change as it should with roll/heave/pitch movements of suspension.
I will report on my progress.
Thank you
Ted
Tudor Miron
Tudor Miron am 19 Mär. 2025
Bearbeitet: Tudor Miron am 21 Mär. 2025
Hi Jack,
Unfortunately, I have to admit that I wasn't able to succesfully implement strategies that you suggested. If I use CPI tire from VD blockset then tire simply falls through the ground. I just can't figure out how to use it properly. With MF tire block I simply use infinite plane between ground and B port of MF tire block. With CPI tire I can't find a port similar to above. I think that it requires something like external force/torque block that it will feed and which will transmit tire forces from tire back to the hub and ground surface but don't know how to implement it properly. May I ask you to give me a hint on how should I do it?
I tryed adding constrain blocks - planar joint. Added high values to upper/lower limits but it doesn't seem to work - model keeps sliding away.
Slowing the solver (I'm using ode23t) didn't make any difference (when still using MF tire block).
Thank you,
Ted

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (0)

Kategorien

Mehr zu Brakes and Detents finden Sie in Help Center und File Exchange

Produkte


Version

R2023a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by