# pollen-robotics/microduck_rl

> Reinforcement learning environments for the Microduck bipedal robot built on mjlab (MuJoCo Warp) with PPO: train walking and trick policies, export to ONNX, and deploy to the real robot.

- Magnitude: 1.7 out of 10 — Steady
- Stars: 2,288 total · +7 stars today, ≈ 10 by evening
- Star trust: star growth looks organic
- Category: Hardware, IoT & robotics · Language: Python · License: Apache-2.0 · Created: 2025-12-06 · Last push: 2026-09-26
- GitHub: https://github.com/pollen-robotics/microduck_rl · Homepage: https://pollen-robotics.com/microduck · Page: https://gitnova.dev/en/r/pollen-robotics/microduck_rl

## Useful for

- Train a Microduck walking policy with velocity tracking on GPU
- Export a trained policy to ONNX for deployment on the real robot
- Test a policy in simulation with keyboard control and CSV logging

## Why it’s here

- 7 stars so far today, about 10 expected by the end of the day.
- GitHub Trending this month: #9, +2,282 stars.
- Recent forks include notable developers: @zhangziliang04 (277 followers).

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 494
- Issues and pull requests: 61
- Watchers: 12
- Average over the last week: 11 per day
- Usual pace: 41 per day
- Stars in the last hour (measured): 2

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-08-29 … 2026-09-27: 155, 312, 361, 196, 135, 91, 50, 36, 65, 76, 54, 30, 39, 23, 16, 20, 30, 14, 15, 19, 12, 8, 19, 18, 6, 18, 13, 9, 5, 7

## Spotted in now

- GitHub Trending this month: #9, +2,282 stars

## Similar by description

1. **AI-FanGe/Microduck-build-tutorial** — 1.2 · Cooling · Hardware, IoT & robotics · Python · +4 stars today, ≈ 6 by evening
   Microduck is a compact bipedal robot project on Raspberry Pi Zero 2 W with 14 Dynamixel XL330 servos and a BNO08x IMU. It includes deployment code for walking via an ONNX policy and a reinforcement learning training environment based on…
   Full card: https://gitnova.dev/en/r/AI-FanGe/Microduck-build-tutorial.md
2. **zjwzcx/Awesome-Astra-Embodied-AI** — 1.3 · Cooling · Learning & lists · +2 stars today, ≈ 3 by evening
   A curated collection of demos and cases of GPT-6 Astra in embodied AI and robotics: zero-shot control, deployment in simulation and on real robots, RL training, and real-to-sim replay.
   Full card: https://gitnova.dev/en/r/zjwzcx/Awesome-Astra-Embodied-AI.md
3. **pollen-robotics/microduck** — 2.2 · Steady · Hardware, IoT & robotics · Rust · +13 stars today, ≈ 20 by evening
   Firmware and system software for Microduck, a tiny biped duck robot on a Rockchip RK3566 that walks, rolls and gets up using neural policies. The repo contains daemons for motor control, updates, Bluetooth, gamepad and camera.
   Full card: https://gitnova.dev/en/r/pollen-robotics/microduck.md
4. **jianglongye/dexterous-astra** — 0.0 · Steady · Hardware, IoT & robotics · Python · +0 stars today, ≈ 0 by evening
   Training artifacts from experiments in dexterous robot manipulation: simulated environments, zero-shot scripts, and trained checkpoints for pen spinning and Rubik's cube solving.
   Full card: https://gitnova.dev/en/r/jianglongye/dexterous-astra.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-09-27 16:22 UTC, updated every 30 minutes.