juicenv/microduck-tug-anchor1

tug-anchor1

Tug of War anchor: fine-tuned from Nico-robot/microduck-tug-of-war (via our tug-puller line) against a league of frozen opponent ducks, trained to squat and lean back. Plain network, no seed tables.

A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs until told otherwise — a gait, loaded into a policy slot.

Run it on a robot

sudo robotctl policy load <slot> juicenv/microduck-tug-anchor1

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 Nico-robot/microduck-tug-of-war.

Training

• repo: pollen-robotics/microduck_rl
• branch: HEAD
• commit: 5cee942d8