task.xml

<mujoco model="Cart-Pole Swing-Up">
  <include file="../common.xml"/>
  <!-- modified from: https://github.com/google-deepmind/dm_control/blob/main/dm_control/suite/cartpole.xml -->
  <include file="cartpole_modified.xml" />

  <size memory="4K"/>

  <custom>
    <!-- agent -->
    <numeric name="agent_planner" data="1" />
    <numeric name="agent_horizon" data="1.0" />
    <numeric name="agent_timestep" data="0.01" />
    <numeric name="sampling_sample_width" data="0.01" />
    <numeric name="sampling_control_width" data="0.015" />
    <numeric name="sampling_spline_points" data="10" />
    <numeric name="sampling_exploration" data="0.5" />
    <numeric name="gradient_spline_points" data="10" />
    <numeric name="residual_Goal" data="0.0 -1.5 1.5" />

    <!-- estimator -->
    <numeric name="estimator" data="0" />
    <numeric name="estimator_sensor_start" data="4" />
    <numeric name="estimator_number_sensor" data="2" />
    <numeric name="estimator_timestep" data="0.005" />
    <numeric name="batch_configuration_length" data="3" />
    <numeric name="batch_scale_prior" data="1.0e-3" />
  </custom>

  <sensor>
    <!-- cost -->
    <user name="Vertical" dim="1" user="6 10.0 0 100.0 0.01"/>
    <user name="Centered" dim="1" user="6 10.0 0 100.0 0.1"/>
    <user name="Velocity" dim="1" user="0 0.1 0.0 1.0"/>
    <user name="Control" dim="1" user="0 0.1 0.0 1.0"/>

    <!-- estimator -->
    <jointpos name="slider_pos" joint="slider"/>
    <jointpos name="hinge_pos" joint="hinge_1"/>

    <!-- trace -->
    <framepos name="trace0" objtype="site" objname="tip"/>

    <!-- residual -->
    <framepos name="position" objtype="site" objname="tip"/>
    <framelinvel name="velocity" objtype="site" objname="tip"/>
  </sensor>
  <keyframe>
    <key name="home" qpos="1 0" />
  </keyframe>
</mujoco>