Microduck skill tree (sim only)
Collection
Policies trained in simulation for a Pollen Robotics Microduck still on order, one level per skill, plus the training log. Sim only so far. • 6 items • Updated
walk-flat
Flat-ground walking gait (Mjlab-Velocity-Flat-MicroDuck), trained in simulation only: 4,000 iterations, 4096 envs, 58 min on one RTX 5090. In Pollen's inference sim it held 0.15 m/s on a 0.30 m/s command for 12 s, trunk at 119 to 121 mm. Not yet tested on a real Microduck.
A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise — a gait for the walk slot.
Run it on a robot
sudo robotctl policy load walk witcheer/microduck-walk-flat
The observation normalizer is baked into policy.onnx; feed raw observations.
manifest.json follows schema 2 of the microduck policy manifest (docs/policy-manifest.md in the daemon repo).
Training
- repo:
pollen-robotics/microduck_rl - branch:
develop - commit:
53b8971b6