State machine: - sit/stand hand-over timers are now cancellable and cleared by every transition entry point (setMode, triggerRoll, resetSim); a stale 2 s stand-up timer could previously flip a sitting or rolling duck to the walking policy mid-motion - resetSim cancels any roll in flight and pending timers, always lands on walk (R mid-roll no longer replays the roll on the reset duck) - rolls only launch from a standing walk, not from a sit - RT quack uses a Schmitt trigger (fire at 0.35, re-arm below 0.2): analog jitter around a single threshold caused multiple quacks Sound: - ship the runtime's four chirp voice banks (12 takes each); each colourway has its own duck voice and every quack picks a random take Co-authored-by: Cursor <cursoragent@cursor.com>
title, emoji, colorFrom, colorTo, sdk, pinned
| title | emoji | colorFrom | colorTo | sdk | pinned |
|---|---|---|---|---|---|
| Microduck Sandbox | 🐤 | yellow | gray | static | false |
Microduck RL playground
Real trained RL policies for the Microduck robot, running fully in the browser: MuJoCo compiled to WebAssembly steps the physics, onnxruntime-web runs the policy network at 50 Hz. No server, no backend.
Policies
| Button | Checkpoint | What it does |
|---|---|---|
| Run | BEST_alpha_walking.onnx |
Velocity-tracking locomotion (arrows / WASD to steer) |
| Sit | BEST_alpha_sitstand.onnx |
Sits down on its hull, stands back up |
| Roll | roulade.onnx |
Rolls over and recovers |
Policies and MJCF model from apirrone/microduck_runtime and apirrone/mjlab_microduck.
Controls
- Arrow up / down: forward / back
- Arrow left / right: turn
- A / E: strafe
- Space: roll (one barrel roll, then back to running)
- R: reset
- Drag to orbit, scroll to zoom
- Colour dots: repaint the duck (it quacks)
Gamepad
Plug in a controller and the same mapping as the real robot runtime applies:
- Left stick: forward / back + strafe
- Right stick: turn
- X: roll
- D-pad down: sit / stand toggle
- Right trigger: mouth (analog) + quack
Multiplayer ghosts
Other people visiting the Space at the same time show up as translucent ducks, live. Peer-to-peer WebRTC via Trystero (serverless signaling over public Nostr relays), so it works from a static Space with no backend. Each tab broadcasts its duck's pose (trunk + 14 joints + jaw + colour) at 10 Hz; up to 3 ghosts are rendered, extra peers stay connected but invisible. The header shows "N online" when peers are around.
How it works
rl.jsfetches the MJCF (robot_allcollisions.xml), strips the visual geoms, injects a floor and a STAND keyframe, and compiles it with the official@mujoco/mujocoWASM bindings.- The 61D observation (gyro, projected gravity, joint pos/vel, last action,
command) matches
mjlab_microduck/scripts/infer_policy.py. - Rendering is a three.js rig built from
kinematics.json+ decimated STL meshes, driven directly from MuJoCoqpos.