Tendon and Inertia #2514
Replies: 1 comment 3 replies
-
|
Nice that you're asking for this, we are currently working on this feature 🙂 Hopefully in the next month or so. |
Beta Was this translation helpful? Give feedback.
-
|
This sounds great! Please give an update then the feature is released! 🙂 |
Beta Was this translation helpful? Give feedback.
-
|
Out of interest, do you think it be a valid way to model this (for the time being) by adding a dummy body on the actuated end, attaching the tendon to a site on it, and adding an actuated vertical slide joint to that body (with configurable armature, passive properties, etc)? Example with dummy body<mujoco model="my_cable_robot">
<compiler angle="radian"/>
<asset>
<texture type="skybox" builtin="gradient" rgb1="1 1 1" rgb2="0.6 0.8 1" width="256" height="1536"/>
<texture type="2d" name="texplane" builtin="checker" mark="cross" rgb1="0.2 0.3 0.4" rgb2="0.1 0.15 0.2" markrgb="0.8 0.8 0.8" width="512" height="512"/>
<material name="matplane" texuniform="true" texrepeat="3 3" reflectance="0.1"/>
</asset>
<worldbody>
<body name="end_effector" pos="2.183 0.16 1">
<joint name="z_end_effector" pos="0 0 0" axis="0 0 1" type="slide"/>
<geom name="end_effector" size="0.01 0.01 0.01" type="box" mass="0.5" rgba="1 0 0 1"/>
<site name="FTR_end_effector" pos="0 0 0"/>
</body>
<body name="FTR_module_assembly" pos="2.223 0.16 2.44">
<body pos="0.02 0 -0.5">
<site name="FTR_cable_fix_point" />
<inertial mass="1e-5" pos="0 0 0" diaginertia="1e-10 1e-10 1e-10"/>
<joint name="slide_j" type="slide" axis="0 0 1" armature="0.1" range="-2 0" limited="true" damping="0.1" solimplimit="0.999 0.999 0.00001 0.5 2" solreflimit="0.0002 1"/>
</body>
<body name="FTR_pulley" quat="0.707107 -0.707107 0 0">
<joint name="FTR_roller" pos="0 0 0" axis="0 1 0" range="-1.5708 1.5708"/>
<geom name="FTR_roller" size="0.02 0.02" pos="-0.02 0 0" type="cylinder" rgba="1 1 1 1"/>
<site name="FTR_roller_fix" pos="-0.02 -0.021 0"/>
</body>
</body>
</worldbody>
<tendon>
<spatial name="FTR_tendon" limited="true" range="0 3.5" rgba="0 1 0 1">
<site site="FTR_cable_fix_point"/>
<geom geom="FTR_roller"/>
<site site="FTR_roller_fix"/>
<geom geom="FTR_roller"/>
<site site="FTR_end_effector"/>
</spatial>
</tendon>
<actuator>
<general name="FTR_cable_force" joint="slide_j" ctrlrange="-10 0"/>
</actuator>
<sensor>
<tendonpos tendon="FTR_tendon" name="FTR_cable_length"/>
<tendonvel tendon="FTR_tendon" name="FTR_cable_vel"/>
</sensor>
</mujoco> |
Beta Was this translation helpful? Give feedback.
-
|
Hi, I think this is possible. It actually did model the mechanism this way, firstly. But I changed it as I like the tidy look with the force directly applied to the tendon. With multiple tendons for the cable driven parallel robot, it gets really messy with your way, but it is good to keep this in mind as backup. Does the small mass introduce any drawbacks in the simulation? And as we are talking about the armature: What is the unit of this parameter? Is it a mass or a force that is always directed against the direction of motion? |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Intro
Hi!
I am a researcher at the University of Augsburg, I use MuJoCo for my research on cable driven parallel robots. In a first step, I want to simulate an Atwood machine.
My setup
I use Python 3.12.5 and MuJoCo 3.2.5.
My question
I am currently simulating an Atwood machine where a mass is attached to a tendon that runs over a pulley. However, instead of having another mass on the other side, the tendon is fixed at one end. I can apply a force F to shorten the tendon and by that moving the mass.
In the real system I want to model, the tendon is wrapped around a motor-driven spool to shorten it. This spool has its own moment of inertia, which I would like to include in my simulation.
I can transform the moment of inertia in a translational mass, but this is mass is affected by gravity, while the moment of inertia is not. Is there a way to add mass to the tendon that is not affected by gravity? Something like armature for joints? Or is there any other way to model this effects? Can I add a cylindrical body next to the tendon and add a constraint that the body has to rotate along with the tendon?
Thank you!
Minimal model and/or code that explain my question
Here is the model of the basic Atwood machine (without inertia):
Model:
minimal XML
Confirmations
Beta Was this translation helpful? Give feedback.
All reactions