Simulation testedlocomotionv0.9.0Apache-2.0

Waddle Custom Locomotion

Independent community-developed walking gait featuring a pronounced rhythmic waddling sway tuned for biped compliance.

Waddle gait simulated in MuJoCo viewer displaying rhythmic sway

About this behavior

Developed by Nick Koenig (nickoenig37) as an exploration of duck-like biomimetic locomotion. Modifies reward terms on lateral torso oscillation and foot contact cadence to produce a distinctive waddle gait while preserving the standard 61-dim observation and 14-actuator contract.

Runtime contract

61-dimension input · 14 joint targets · 50 Hz

Show details

50 Hz / 4×slot: walk

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 actuator physics). The beak motor is separate from this fixed locomotion action space.

Verification

Status: Simulation tested
Evidence: Simulated and validated in mjlab across 4096 parallel environments on multi-GPU cluster with video recording.
Target hardware: MicroDuck v1 / Waddle Custom Biped Platform
Simulation: mjlab (MuJoCo Warp) at 50 Hz
Note: Community policy. Exhibits higher lateral roll oscillation than the official Alpha walk.

Compatibility

Robot model: custom-duck
MJCF model: robot_walk.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: walk

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 waddle_walking.onnx

curl -L "https://raw.githubusercontent.com/nickoenig37/mjlab_microduck_waddle/main/docs/waddle_walking.onnx" -o "/opt/robot/policies/waddle_walking.onnx"

2 Add the policy to robotd

[policy]
walk = "/opt/robot/policies/waddle_walking.onnx"

3 Restart the control loop

sudo systemctl restart robotd
robotctl health

⌁ Try it in simulation

uv run play Mjlab-Velocity-Flat-Waddle

Sources

Upstream repositorynickoenig37/mjlab_microduck_waddle Training codeMjlab-Velocity-Flat-Waddle