Documentation: RobotLocomotion/drake@ee25dbb

Author: drake-jenkins-bot

doxygen_cxx/classdrake_1_1systems_1_1controllers_1_1_inverse_dynamics.html

@@ -178,7 +178,7 @@
 </pre><p> where <code>M(q)</code> is the mass matrix, <code>C(q, v)v</code> is the bias term containing Coriolis and gyroscopic effects, <code>τ_g(q)</code> is the vector of generalized forces due to gravity and <code>τ_app</code> contains applied forces from force elements added to the multibody model (this can include damping, springs, etc. See MultibodyPlant::CalcForceElementsContribution()).</p>
 <p>The system also provides a pure gravity compensation mode via an option in the constructor. In this case, the output is simply </p><pre>
  τ_id = -τ_g(q).
-</pre><dl class="section note"><dt>Note</dt><dd>As an alternative to adding a controller to your diagram, gravity compensation can be modeled by disabling gravity for a given model instance, see MultibodyPlant::set_gravity_enabled().</dd></dl>
+</pre><dl class="section note"><dt>Note</dt><dd>As an alternative to adding a controller to your diagram, gravity compensation can be modeled by disabling gravity for a given model instance, see MultibodyPlant::set_gravity_enabled(), unless the gravity compensation needs to be accounted for when evaluating effort limits.</dd></dl>
 <p><a class="el" href="classdrake_1_1systems_1_1controllers_1_1_inverse_dynamics_controller.html" title="A state feedback controller that uses a PidController to generate desired accelerations,...">InverseDynamicsController</a> uses a PID controller to generate desired acceleration and uses this class to compute generalized forces. Use this class directly if desired acceleration is computed differently.</p>
 <table align="center" cellpadding="0" cellspacing="0">
 <tr align="center">

doxygen_cxx/classdrake_1_1systems_1_1controllers_1_1_inverse_dynamics_controller.html

@@ -193,7 +193,6 @@
 </table>
 </td></tr>
 </table>
-<dl class="section note"><dt>Note</dt><dd>As an alternative to adding a separate controller system to your diagram, you can model gravity compensation with PD controllers using MultibodyPlant APIs. Refer to MultibodyPlant::set_gravity_enabled() as an alternative to modeling gravity compensation. To model PD controlled actuators, refer to <a class="el" href="classdrake_1_1multibody_1_1_multibody_plant.html#mbp_actuation">Actuation</a>.</dd></dl>
 <p>The desired acceleration port shown in <span style="color:gray">gray</span> may be absent, depending on the arguments passed to the constructor.</p>
 <p>Note that this class assumes the robot is fully actuated, its position and velocity have the same dimension, and it does not have a floating base. If violated, the program will abort. This controller was not designed for closed-loop systems: the controller accounts for neither constraint forces nor actuator forces applied at loop constraints. Use on such systems is not recommended.</p>
 <dl class="section see"><dt>See also</dt><dd><a class="el" href="classdrake_1_1systems_1_1controllers_1_1_inverse_dynamics.html" title="Solves inverse dynamics with no consideration for joint actuator force limits.">InverseDynamics</a> for an accounting of <a class="el" href="namespacedrake.html#adcc2d90ce6051e150effadd72fac548c" title="Checks truth for all elements in matrix m.">all</a> forces incorporated into the inverse dynamics computation.</dd></dl>

doxygen_cxx/classdrake_1_1systems_1_1controllers_1_1_joint_stiffness_controller.html

@@ -193,6 +193,7 @@
 </td></tr>
 </table>
 <p>Note that the joint impedance control as implemented on Kuka's iiwa and Franka' Panda is best modeled as a stiffness controller (unless one were to model the actual elastic joints and rotor-inertia shaping). See <a href="https://manipulation.csail.mit.edu/force.html">https://manipulation.csail.mit.edu/force.html</a> for more details.</p>
+<dl class="section note"><dt>Note</dt><dd>As an alternative to adding a separate controller system to your diagram, you can model a JointStiffness controller using MultibodyPlant APIs. Refer to MultibodyPlant::set_gravity_enabled() as an alternative to modeling gravity compensation (unless the gravity compensation terms need to be accounted for when computing effort limits). To model PD controlled actuators, refer to <a class="el" href="classdrake_1_1multibody_1_1_multibody_plant.html#mbp_actuation">Actuation</a>.</dd></dl>
 <dl class="tparams"><dt>Template Parameters</dt><dd>
   <table class="tparams">
     <tr><td class="paramname">T</td><td>The scalar type, which must be one of the <a class="el" href="group__default__scalars.html">default scalars</a>. </td></tr>

pydrake/pydrake.systems.controllers.html

@@ -551,7 +551,9 @@
 <p class="admonition-title">Note</p>
 <p>As an alternative to adding a controller to your diagram, gravity
 compensation can be modeled by disabling gravity for a given model
-instance, see MultibodyPlant::set_gravity_enabled().</p>
+instance, see MultibodyPlant::set_gravity_enabled(), unless the
+gravity compensation needs to be accounted for when evaluating
+effort limits.</p>
 </div>
 <p>InverseDynamicsController uses a PID controller to generate desired
 acceleration and uses this class to compute generalized forces. Use
@@ -672,16 +674,7 @@
 matrix. The subscript <code class="docutils literal notranslate"><span class="pre">_d</span></code> indicates desired values, and
 <code class="docutils literal notranslate"><span class="pre">vd_command</span></code> indicates the acceleration command (which includes the
 stabilization terms) passed to the inverse dynamics computation.</p>
-<table align=center cellpadding=0 cellspacing=0><tr align=center><td style="vertical-align:middle"><table cellspacing=0 cellpadding=0><tr><td align=right style="padding:5px 0px 5px 0px">estimated_state&rarr;</td></tr><tr><td align=right style="padding:5px 0px 5px 0px">desired_state&rarr;</td></tr><tr><td align=right style="padding:5px 0px 5px 0px"><span style="color:gray">desired_acceleration</span>&rarr;</td></tr></table></td><td align=center style="border:solid;padding-left:20px;padding-right:20px;vertical-align:middle" bgcolor=#F0F0F0>InverseDynamicsController</td><td style="vertical-align:middle"><table cellspacing=0 cellpadding=0><tr><td align=left style="padding:5px 0px 5px 0px">&rarr; actuation</td></tr><tr><td align=left style="padding:5px 0px 5px 0px">&rarr; generalized_force</td></tr></table></td></tr></table><div class="admonition note">
-<p class="admonition-title">Note</p>
-<p>As an alternative to adding a separate controller system to your
-diagram, you can model gravity compensation with PD controllers
-using MultibodyPlant APIs. Refer to
-MultibodyPlant::set_gravity_enabled() as an alternative to
-modeling gravity compensation. To model PD controlled actuators,
-refer to mbp_actuation “Actuation”.</p>
-</div>
-<p>The desired acceleration port shown in &lt;span
+<table align=center cellpadding=0 cellspacing=0><tr align=center><td style="vertical-align:middle"><table cellspacing=0 cellpadding=0><tr><td align=right style="padding:5px 0px 5px 0px">estimated_state&rarr;</td></tr><tr><td align=right style="padding:5px 0px 5px 0px">desired_state&rarr;</td></tr><tr><td align=right style="padding:5px 0px 5px 0px"><span style="color:gray">desired_acceleration</span>&rarr;</td></tr></table></td><td align=center style="border:solid;padding-left:20px;padding-right:20px;vertical-align:middle" bgcolor=#F0F0F0>InverseDynamicsController</td><td style="vertical-align:middle"><table cellspacing=0 cellpadding=0><tr><td align=left style="padding:5px 0px 5px 0px">&rarr; actuation</td></tr><tr><td align=left style="padding:5px 0px 5px 0px">&rarr; generalized_force</td></tr></table></td></tr></table><p>The desired acceleration port shown in &lt;span
 style=”color:gray”&gt;gray&lt;/span&gt; may be absent, depending on the
 arguments passed to the constructor.</p>
 <p>Note that this class assumes the robot is fully actuated, its position
@@ -816,6 +809,16 @@
 one were to model the actual elastic joints and rotor-inertia
 shaping). See <a class="reference external" href="https://manipulation.csail.mit.edu/force.html">https://manipulation.csail.mit.edu/force.html</a> for more
 details.</p>
+<div class="admonition note">
+<p class="admonition-title">Note</p>
+<p>As an alternative to adding a separate controller system to your
+diagram, you can model a JointStiffness controller using
+MultibodyPlant APIs. Refer to
+MultibodyPlant::set_gravity_enabled() as an alternative to
+modeling gravity compensation (unless the gravity compensation
+terms need to be accounted for when computing effort limits). To
+model PD controlled actuators, refer to mbp_actuation “Actuation”.</p>
+</div>
 <dl class="py method">
 <dt class="sig sig-object py" id="pydrake.systems.controllers.JointStiffnessController.__init__">
 <span class="sig-name descname"><span class="pre">__init__</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">self</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="#pydrake.systems.controllers.JointStiffnessController" title="pydrake.systems.controllers.JointStiffnessController"><span class="pre">pydrake.systems.controllers.JointStiffnessController</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">plant</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><a class="reference internal" href="pydrake.multibody.plant.html#pydrake.multibody.plant.MultibodyPlant" title="pydrake.multibody.plant.MultibodyPlant"><span class="pre">pydrake.multibody.plant.MultibodyPlant</span></a></span></em>, <em class="sig-param"><span class="n"><span class="pre">kp</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">numpy.ndarray</span><span class="p"><span class="pre">[</span></span><span class="pre">numpy.float64</span><span class="p"><span class="pre">[</span></span><span class="pre">m</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="m"><span class="pre">1</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">kd</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">numpy.ndarray</span><span class="p"><span class="pre">[</span></span><span class="pre">numpy.float64</span><span class="p"><span class="pre">[</span></span><span class="pre">m</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="m"><span class="pre">1</span></span><span class="p"><span class="pre">]</span></span><span class="p"><span class="pre">]</span></span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">None</span></span></span><a class="headerlink" href="#pydrake.systems.controllers.JointStiffnessController.__init__" title="Permalink to this definition"></a></dt>