uDuck registrybehaviors for Microduck
Back to behaviors
Hardware verifiedlocomotionv1.0.0Apache-2.0

Smooth Sit ↔ Stand

Controlled bi-directional transition between upright standing and rested sitting posture with continuous head commandability.

By Pollen Robotics @pollen-robotics

About this behavior

A unified policy trained in MuJoCo to handle both gentle sitting down and standing back up. Softens motor gains during transition to ensure the duck does not slam into the ground. While sitting, head orientation remains commandable through the 4-dim head pose observation slot.

Runtime contract

61-dimension input · 14 joint targets · 50 Hz

Show details
50 Hz / 4×slot: sitstand
Observation vector61 dimensions · normalizer baked in
Proprioception48D

Positions, velocities, previous actions, gravity, and gyro.

Twist command3D

Forward, lateral, and yaw velocity intent.

Head pose4D

Neck and head orientation targets.

Body pose6D

Six-axis torso offset command.

Action targetsscale: 1×
Left leg5 joints

hip yaw · hip roll · hip pitch · knee · ankle

Neck & head4 joints

neck pitch · head pitch · yaw · roll

Right leg5 joints

hip yaw · hip roll · hip pitch · knee · ankle

Actuator model: Dynamixel XL330 (BAM M6 voltage control law). The beak motor is separate from this fixed locomotion action space.

Verification

Status
Hardware verified
Evidence
Verified on physical MicroDuck hardware with gamepad Y button toggle. Robot smoothly sits down and stands back up.
Target hardware
MicroDuck v1 (Rockchip RK3566, 14 Dynamixel XL330-M077)
Simulation
mjlab (MuJoCo Warp) at 50 Hz
Note
Transition utilizes posture flags in twist magnitude; rise runs with softened standing gains.

Compatibility

Robot model
microduck-standard
MJCF model
robot_allcollisions.xml
Terrain
flat
Required accessories
None — standard duck setup

Run it on your duck

Use the canonical model and this policy slot to try the behavior locally.

slot: sitstand
Try the simulation first. Hardware testing assumes a matching Microduck, clear space, and a safe surface. uDuck does not guarantee that a community policy is safe to run.
Deployment stepsShow instructions
1 Download the ONNX model BEST_alpha_sitstand.onnx
curl -L "https://huggingface.co/spaces/pollen-robotics/microduck-simulator/resolve/main/app/public/policies/BEST_alpha_sitstand.onnx" -o "/opt/robot/policies/BEST_alpha_sitstand.onnx"
2 Add the policy to robotd
[policy]
sitstand = "/opt/robot/policies/BEST_alpha_sitstand.onnx"
3 Restart the control loop
sudo systemctl restart robotd
robotctl health
⌁ Try it in simulation
uv run scripts/infer_policy.py --sitstand BEST_alpha_sitstand.onnx --new-cmd-obs

Sources

Upstream repositorypollen-robotics/microduckTraining codeMjlab-SitStand-Flat-MicroDuckInteractive simulatorhttps://huggingface.co/spaces/pollen-robotics/microduck-simulator