Downloads · 30 days
0
microduck-standup
witcheer/microduck-standup
microduck-standup is a reinforcement learning model from witcheer. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. It is set up for onnx.
Get-up policy (Mjlab-StandUp-Flat-MicroDuck), simulation only: 25,000 iterations (the maker's 15,000 plus a 10,000 extension), 4096 envs, 407 min on one RTX 5090. Stands from sitting and from face down and holds stand…
Access
Public
Updated Sep 21, 2026
Repo size
794 KB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx794 KB · 100%
At a glance
- Task
- Reinforcement Learning
- Library
- onnx
- Access
- Public
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026
- SHA
- abb8895c
Model card
From the Hugging Face model README
standup
Get-up policy (Mjlab-StandUp-Flat-MicroDuck), simulation only: 25,000 iterations (the maker's 15,000 plus a 10,000 extension), 4096 envs, 407 min on one RTX 5090. Stands from sitting and from face down and holds standing; from flat on its back it freezes half-rolled (0 of 2 takes). Not yet tested on a real Microduck.
A episodic policy for the microduck (61-D observation, 14 actions, 50 Hz). Runs 6.0 s and returns itself to a standing pose.
Run it on a robot
sudo robotctl policy add standup witcheer/microduck-standup
robotctl robot do standup
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:
develop - commit:
53b8971b6
- Task
- Reinforcement Learning
- Library
- onnx
- Created
- Sep 21, 2026
- Updated
- Sep 21, 2026