Instructions to use linkTheWorld/microduck-duck-bowling with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use linkTheWorld/microduck-duck-bowling with Microduck:
sudo robotctl policy load walk linkTheWorld/microduck-duck-bowling
- Notebooks
- Google Colab
- Kaggle
duck-bowling
Duck Bowling: the roller skater is bowled first, cannot brake and rolls into the gutter; then the sprinter winds up to In the Hall of the Mountain King (Grieg) and charges the pins itself. Four down, a sad trombone, two pins gloat, and the fallen sing Ode to Joy (Beethoven) in three parts lying down. All eight ducks are our own policies: a 2 m Sprint walker (PPO with a linear forward-speed reward) and the 2 m Roller Sprint skater. Simulation only.
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 linkTheWorld/microduck-duck-bowling
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:
sprint-race - commit:
822c9db39
- Downloads last month
- -