chenlin/Mujoco_WASM
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Cosmetic improvements to Actuation Model documentation.

Browse Source 
PiperOrigin-RevId: 526659426
Change-Id: If0ca9a505bb77e496739938a946ead5780b93d0f
This commit is contained in:
Yuval Tassa
2023-04-24 09:02:44 -07:00
committed by Copybara-Service
parent 62a675ba28
commit 51126caf3f
1 changed files with 49 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/computation.rst
+49 -37
View File
@@ -258,7 +258,7 @@ actuator works. The user can set them independently for maximum flexibility, or
.. _geTransmission: .. _geTransmission:
Transmission Transmission
~~~~~~~~~~~~ ^^^^^^^^^^^^
Each actuator has a scalar length :math:`l_i(q)` defined by the type of transmission and its parameters. The gradient Each actuator has a scalar length :math:`l_i(q)` defined by the type of transmission and its parameters. The gradient
:math:`\nabla l_i` is an :math:`n_V`-dimensional vector of moment arms. It determines the mapping from scalar :math:`\nabla l_i` is an :math:`n_V`-dimensional vector of moment arms. It determines the mapping from scalar
@@ -266,19 +266,23 @@ actuator force to joint force. The transmission properties are determined by the
is attached; the possible attachment object types are :at:`joint`, :at:`tendon`, :at:`jointinparent`, is attached; the possible attachment object types are :at:`joint`, :at:`tendon`, :at:`jointinparent`,
:at:`slider-crank`, :at:`site`, and :at:`body`. :at:`slider-crank`, :at:`site`, and :at:`body`.
:at:`joint` and :at:`tendon`
The :at:`joint` and :at:`tendon` transmission types act as expected and correspond to the actuator applying forces or The :at:`joint` and :at:`tendon` transmission types act as expected and correspond to the actuator applying forces or
torques to the target object. Ball joints are special, see the :at:`joint` documentation in torques to the target object. Ball joints are special, see the :ref:`actuator/general/joint<actuator-general-joint>`
:ref:`actuator<actuator-general>` reference for more details. documentation for more details.
:at:`jointinparent`
The :at:`jointinparent` transmission is unique to ball and free joint and asserts that rotation should be measured The :at:`jointinparent` transmission is unique to ball and free joint and asserts that rotation should be measured
in the parent rather than child frame. in the parent rather than child frame.
:at:`slider-crank`
:at:`slider-crank` `transmissions <https://en.wikipedia.org/wiki/Slider-crank_linkage>`_ transform a linear force to :at:`slider-crank` `transmissions <https://en.wikipedia.org/wiki/Slider-crank_linkage>`_ transform a linear force to
a torque, as in a piston-driven combustion engine. `This model a torque, as in a piston-driven combustion engine. `This model
<https://github.com/deepmind/mujoco/tree/main/model/slider_crank>`_ contains pedagogical examples. Slider-cranks can <https://github.com/deepmind/mujoco/tree/main/model/slider_crank>`_ contains pedagogical examples. Slider-cranks can
also be modeled explicitly by creating MuJoCo bodies and coupling them with equality constraints to the rest of the also be modeled explicitly by creating MuJoCo bodies and coupling them with equality constraints to the rest of the
system, but that would be less efficient. system, but that would be less efficient.
:at:`site`
:at:`site` transmission (without a :at:`refsite`, see below) and :at:`body` transmission targets have a fixed zero :at:`site` transmission (without a :at:`refsite`, see below) and :at:`body` transmission targets have a fixed zero
length :math:`l_i(q) = 0`. They can therefore not be used to maintain a desired length, but can be used to apply length :math:`l_i(q) = 0`. They can therefore not be used to maintain a desired length, but can be used to apply
forces. Site transmissions correspond to applying a Cartsian force/torque at the site, and are useful for modeling forces. Site transmissions correspond to applying a Cartsian force/torque at the site, and are useful for modeling
@@ -292,51 +296,59 @@ is attached; the possible attachment object types are :at:`joint`, :at:`tendon`,
be controlled with a :el:`position` actuator, enabling Cartesian end-effector control. See the :at:`refsite` be controlled with a :el:`position` actuator, enabling Cartesian end-effector control. See the :at:`refsite`
documentation in :ref:`actuator<actuator-general>` reference for more details. documentation in :ref:`actuator<actuator-general>` reference for more details.
.. _geActivation:
Activation dynamics Activation dynamics
Some actuators such as pneumatic and hydraulic cylinders as well as biological muscles have an internal state called ^^^^^^^^^^^^^^^^^^^
"activation". This is a true dynamic state, beyond the joint positions :math:`q` and velocities :math:`v`. Including
such actuators in the model results in 3rd-order dynamics. We denote the vector of actuator activations :math:`w`.
They have some first-order dynamics
.. math:: Some actuators such as pneumatic and hydraulic cylinders as well as biological muscles have an internal state called
\dot{w}_i \left( u_i, w_i, l_i, \dot{l}_i \right) "activation". This is a true dynamic state, beyond the joint positions :math:`q` and velocities :math:`v`. Including
such actuators in the model results in 3rd-order dynamics. We denote the vector of actuator activations :math:`w`.
They have some first-order dynamics
determined by the activation type and corresponding model parameters. Note that each actuator has scalar dynamics .. math::
independent of the other actuators. The activation types currently implemented are \dot{w}_i \left( u_i, w_i, l_i, \dot{l}_i \right)
.. math:: determined by the activation type and corresponding model parameters. Note that each actuator has scalar dynamics
\begin{aligned} independent of the other actuators. The activation types currently implemented are
\text{integrator}: & & \dot{w}_i &= u_i \\
\text{filter}: & & \dot{w}_i &= (u_i - w_i) / t \\
\end{aligned}
where :math:`t` is an actuator-specific time constant stored in ``mjModel.actuator_dynprm``. In addition the type can .. math::
be "user", in which case :math:`w_i` is computed by the user-defined callback :ref:`mjcb_act_dyn`. The type can also \begin{aligned}
be "none" which corresponds to a regular actuator with no activation state. The dimensionality of :math:`w` equals \text{integrator}: & & \dot{w}_i &= u_i \\
the number of actuators whose activation type is different from "none". \text{filter}: & & \dot{w}_i &= (u_i - w_i) / t \\
\end{aligned}
where :math:`t` is an actuator-specific time constant stored in ``mjModel.actuator_dynprm``. In addition the type can
be "user", in which case :math:`w_i` is computed by the user-defined callback :ref:`mjcb_act_dyn`. The type can also
be "none" which corresponds to a regular actuator with no activation state. The dimensionality of :math:`w` equals
the number of actuators whose activation type is different from "none".
.. _geActuatorForce:
Force generation Force generation
Each actuator generates a scalar force :math:`p_i` which is some function ^^^^^^^^^^^^^^^^
.. math:: Each actuator generates a scalar force :math:`p_i` which is some function
p_i \left( u_i, w_i, l_i, \dot{l}_i \right)
Similarly to activation dynamics, the force generation mechanism is actuator-specific and cannot interact with the .. math::
other actuators in the model. Currently the force is affine in the activation state when present, and in the control p_i \left( u_i, w_i, l_i, \dot{l}_i \right)
otherwise:
.. math:: Similarly to activation dynamics, the force generation mechanism is actuator-specific and cannot interact with the
p_i = (a w_i \; \text{or} \; a u_i) + b_0 + b_1 l_i + b_2 \dot{l}_i other actuators in the model. Currently the force is affine in the activation state when present, and in the control
otherwise:
Here :math:`a` is an actuator-specific gain parameter and :math:`b_0, b_1, b_2` are actuator-specific bias .. math::
parameters, stored in ``mjModel.actuator_gainprm`` and ``mjModel.actuator_biasprm`` respectively. Different settings p_i = (a w_i \; \text{or} \; a u_i) + b_0 + b_1 l_i + b_2 \dot{l}_i
of the gain and bias parameters can be used to model direct force control as well as position and velocity servos -in
which case the control/activation has the meaning of reference position or velocity. One can also compute custom gain Here :math:`a` is an actuator-specific gain parameter and :math:`b_0, b_1, b_2` are actuator-specific bias
and bias terms by installing the callbacks :ref:`mjcb_act_gain` and :ref:`mjcb_act_bias` and setting the gain and parameters, stored in ``mjModel.actuator_gainprm`` and ``mjModel.actuator_biasprm`` respectively. Different settings
bias type to "user". Note that affine force generation makes it possible to infer the controls/activations from the of the gain and bias parameters can be used to model direct force control as well as position and velocity servos -in
applied force computed in inverse dynamics, using the pseudo-inverse of the matrix of moment arms. However some of which case the control/activation has the meaning of reference position or velocity. One can also compute custom gain
the actuators used in the real world are not affine (especially those that have embedded low-level controllers), so and bias terms by installing the callbacks :ref:`mjcb_act_gain` and :ref:`mjcb_act_bias` and setting the gain and
we are considering extensions to the above model. bias type to "user". Note that affine force generation makes it possible to infer the controls/activations from the
applied force computed in inverse dynamics, using the pseudo-inverse of the matrix of moment arms. However some of
the actuators used in the real world are not affine (especially those that have embedded low-level controllers), so
we are considering extensions to the above model.
Putting all this together, the net force in generalized coordinates contributed by all actuators is Putting all this together, the net force in generalized coordinates contributed by all actuators is