rough-walk-e
Drop-in replacement for alpha_walking fine-tuned on a hostile-terrain ladder of tiles, small stairs, rubble and slopes, with roughly half the falls; the from-scratch rough-walk-g is sturdier on stairs and slopes at 17 percent more motor power. Stated limits include falling on downward steps of 2 cm or more. Sim-only.
https://huggingface.co/RemiFabre/microduck-rough-walk-e61 OBSERVATIONS 14 ACTIONS 50 HZ COMPATIBILITY NOT INDEPENDENTLY VERIFIED
AWESOME MICRODUCK INDEX
rough-walk-e
Drop-in replacement for alpha_walking fine-tuned on a hostile-terrain ladder of tiles, small stairs, rubble and slopes, with roughly half the falls; the from-scratch rough-walk-g is sturdier on stairs and slopes at 17 percent more motor power. Stated limits include falling on downward steps of 2 cm or more. Sim-only.
Project Overview
--- library name: onnx pipeline tag: reinforcement-learning tags: [microduck, microduck-policy, mjlab, robotics] license: apache-2.0 ---
Velocity-commanded walking tuned for rough ground (grid tiles, small stairs, rubble, slopes): the shipped alpha walking fine-tuned 800 iterations on a hostile-terrain ladder. Same command scheme, same motor-power envelope, roughly half the falls.
Command — twist slots of the 13-D command block (head / body slots: 4 head-joint deltas from HOME, rad (neck pitch, head pitch, head yaw, head roll) — tracked, same as alpha walking, unused (zeros)); idle = [0, 0, 0]
Task
| Field | Value |
|---|---|
| Source | HUGGING FACE MODEL SOURCE |
| Runtime contract | 61 observations / 14 actions / 50 Hz |
| Verification | Pending independent Duckio review |
| Artifact | Not indexed |
Intended Use
Use this entry to assess the project and locate its canonical implementation. Read the complete README, inspect the manifest and artifacts, and reproduce the source's evaluation before assigning simulation or hardware verification.