Best practice for modeling specific Coefficient Of Restitution (e) and Coefficient Of Friction (μ) for a tennis ball and court? #2930

xSpaceCoder · 2025-11-05T08:48:14Z

xSpaceCoder
Nov 5, 2025

Intro

Hi!

I am a graduate student at Augsburg University in Germany, I use MuJoCo for my research on simulating a tennis ball's trajectory where I try to accurately match real-world physical properties.

My setup

MuJoCo: 3.3.2
Python: 3.10.12

My question

From literature, my target values are:

Coefficient of Restitution (COR): e≈0.8
Coefficient of Friction (COF): μ≈0.7

I have already configured the ball's geometry, mass, and fluid parameters (with fluidshape="ellipsoid", drag and lift (magnus) coefficients)

My core difficulty is understanding how to correctly map these two physical coefficients (COR and COF) to MuJoCo's contact and solver parameters. Here is how I choose the values of the contact as of now:
solref:

0: stiffness of ball
1: damping of ball

friction:

0&1: COF (no anisotropic behavior expected)
2: default
3&4: rolling coefficient (no anisotropic behavior expected)

solreffriction

0&1: ???

So my question is how do I need to adapt the solreffriction values (and maybe also further values) in order to simulate an accurate tennis ball trajecory?

Minimal model and/or code that explain my question

Here is my reduced XML model:

Model:

<mujoco>
  <option cone="elliptic" gravity="0 0 -9.81" integrator="implicitfast" timestep="{TIMESTEP}" density="1.225" viscosity="0.00001789" />
  <asset>
    <material name="ball_material" reflectance="0" rgba="1 1 1 1"/>
    <texture name="court_texture" type="2d" file="{COURT_ASSET_PATH}" />
    <material name="court_material" reflectance="0.05" texture="court_texture" texrepeat="1 1" texuniform="true"/>
  </asset>
  <worldbody>
    <geom name="floor_geom" type="plane" size="800 800 0.1" rgba="0.6 0.6 0.6 1" pos="0 0 -0.1" material="ball_material"/>
    <geom name="court_plane_geom" type="plane" size="{PPA_LENGTH/2} {PPA_WIDTH/2} 0.1" rgba="0.6 0.6 0.6 1" pos="0 0 0" material="court_material"/>
    <body name="ball_body" pos="0 0 1.2">
      <freejoint/>  
      <geom name="ball_geom" type="sphere" size="{BALL_RADIUS}" material="ball_material" mass="{BALL_MASS}" fluidshape="ellipsoid" fluidcoef="0.66 0.66 0.0 1.0 0.3"/>
    </body>
  </worldbody>
  <default>
    <pair solref="-20000 -21.52" solreffriction="0.0 -100.0" friction="0.7 0.7 0.005 0.04 0.04" solimp="0.98 0.99 0.001 0.5 2"/>
  </default>
  <contact>
    <pair geom1="ball_geom" geom2="court_plane_geom"/>
  </contact>
</mujoco>

Confirmations

I searched the latest documentation thoroughly before posting.
I searched previous Issues and Discussions, I am certain this has not been raised before.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Best practice for modeling specific Coefficient Of Restitution (e) and Coefficient Of Friction (μ) for a tennis ball and court? #2930

Uh oh!

{{title}}

Uh oh!

Replies: 0 comments

Select a reply

Uh oh!

Best practice for modeling specific Coefficient Of Restitution (e) and Coefficient Of Friction (μ) for a tennis ball and court? #2930

Uh oh!

xSpaceCoder Nov 5, 2025

Intro

My setup

My question

Minimal model and/or code that explain my question

Confirmations

Replies: 0 comments

xSpaceCoder
Nov 5, 2025