got-a-talent-reference
Sits and stands on the sit flag, head on command; checkpoint 2000, the MicroDuck Got a Talent reference.
A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise — a gait for the sitstand slot.
Run it on a robot
sudo robotctl policy load sitstand pollen-robotics/microduck-got-a-talent-reference
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
- task_id:
Mjlab-GotATalent-MicroDuck - repo:
https://github.com/pollen-robotics/microduck-challenges.git - branch:
swag-stage - commit:
0f11bbdd9 - checkpoint:
2000 - seed:
1 - base:
mjlab-microduck 0.1.0 @ 981a279c6 - started:
2026-09-29T13:22:42Z
Reproduce
Same code, same uv.lock, same command, same seed. Training it again yields a comparable policy, not the same weights: GPU reinforcement learning is not bit-reproducible across machines.
git clone https://github.com/pollen-robotics/microduck-challenges.git
cd microduck-challenges
git checkout 0f11bbdd9
uv sync
uv run train Mjlab-SwagContest-MicroDuck --env.scene.num-envs 4096 --agent.max-iterations 3000 --agent.seed 1 --agent.logger tensorboard --agent.run-name reference
- Downloads last month
- 14