Instructions to use dashi/microduck-roller-remix-r900 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use dashi/microduck-roller-remix-r900 with Microduck:
sudo robotctl policy load walk dashi/microduck-roller-remix-r900
- Notebooks
- Google Colab
- Kaggle
roller-remix-r900
2 m Roller Sprint remix: DenRakEiw/microduck-senate_duck_0904 unchanged for seeds 2-5; seed 1 gets one more branch (closed-loop per-step action deltas from policy step 25, searched by CMA-ES on that seed alone, simulation only). Credit to DenRakEiw, austindixson, jrubiosainz, suraj-phanindra, juicenv, namenu and Pollen Robotics.
A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz, on rollers). Runs until told otherwise — a gait for the walk slot.
Run it on a robot
sudo robotctl policy load walk dashi/microduck-roller-remix-r900
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).
A remix of DenRakEiw/microduck-senate_duck_0904.
Training
- repo:
pollen-robotics/microduck_rl - branch:
roll-backward - commit:
83599404a
- Downloads last month
- 27