Official Microduck resources

The official material is spread across a GitHub repository, a product site and a blog. This page collects it and says what each entry actually solves, so you do not have to open all of them to find out.

Code and training

• Microduck GitHub — the main software repository. robotctl, the daemons, the design documents and the contribution path. Any “what does this command actually do” question is answered in the source.
• microduck_rl — the reinforcement learning repository: training environments, PPO configuration, MuJoCo Warp, and the ONNX export the robot loads.

Day to day

• README — repository layout, getting started, training, and the documentation index. Read this first.
• robotctl cheat sheet — every command: status, configuration, updates, logs, service management.
• Pairing a gamepad — the pairing steps and what to try when it will not bond. The button mapping is in the cheat sheet, not here.
• duckctl — the robot from a laptop over Bluetooth, with no network and no ssh.

Development

• How it works — the whole system on one page: the daemons, the bus, how an update reaches a robot.
• Set up a dev board — from a blank board to one that takes branch builds.
• Dev cheat sheet — branch builds, release candidates, and the restart traps after an update.
• Push your branch — build on your machine, install over ssh, about a minute.

Product and simulation

• Product page — pre-order, videos, shipped moves, colourways, specification.
• Launch post — the official introduction, including the design trade-offs and who it is for.
• Online simulator — the Microduck Sandbox on Hugging Face Spaces. No hardware needed.
• Press kit — fact sheet, photos, logos, colours, pricing, regions.