microduck-bow / README.md
dostafan's picture
publish bow: perpetual, bowdepth-trhlil2k.onnx
5bd072c verified
|
Raw History Blame Contribute Delete
900 Bytes
metadata
tags:
  - microduck
  - robotics
  - reinforcement-learning
  - onnx
library_name: microduck
pipeline_tag: robotics

bow

Bow with a plié; depth follows the command (0 stand .. 1 full bow)

A perpetual policy for the microduck (61-D observation, 14 actions, 50 Hz). Holds until told otherwise; the daemon drives command.idle for 1.5 s before handing back to the gait.

Run it on a robot

sudo robotctl policy add bow dostafan/microduck-bow --hold <seconds>
robotctl robot do bow

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: bow
  • commit: 0daab160d
  • exported from a checkout with uncommitted changes