LuwuDynamics/xgoduck_hardware

Build your own robot duck. XGO-Duck is a 3D-printable biped powered by Arduino UNO Q and 15 servos, adapted from Pollen Robotics' Microduck. Includes mechanical models, PCB designs, a BOM, and an assembly guide. Build it, explore its motion, and make it your own.

What it solves

XGO-Duck provides a complete, open-source hardware blueprint for building a bipedal, duck-shaped robot. It bridges the gap between physical construction and AI-driven movement by providing the necessary mechanical designs, electronics specifications, and integration points for reinforcement learning policies.

How it works

The robot is built around an Arduino UNO Q controller and 15 serial servos that manage movement across the legs, neck, and beak. The system uses a custom expansion board to integrate an IMU for sensor feedback. The hardware is designed to run motion policies (exported as ONNX files) that are trained in a separate simulation environment, allowing the physical robot to execute complex behaviors like walking or performing a somersault.

Who it’s for

This project is for makers, robotics hobbyists, and researchers interested in embodied intelligence. It is ideal for those who want to experiment with reinforcement learning on physical hardware, as it provides a standardized platform to test and deploy trained motion policies.

Highlights

  • Complete Hardware Kit: Includes 3D-printable STL files, a detailed bill of materials (BOM), and custom expansion board schematics.
  • Reinforcement Learning Ready: Designed to work with companion repositories for training policies in simulation and deploying them via ONNX.
  • High Articulation: Features 15 degrees of freedom, including a movable head and beak, powered by serial servos.
  • Modular Ecosystem: Part of a broader framework that separates hardware assembly, onboard runtime, and AI training into distinct, manageable repositories.

Related

  • Project
    AI-FanGe/Microduck-build-tutorial

    A compact biped robot project providing a full pipeline from MuJoCo-based RL training to real-world deployment on a Raspberry Pi Zero 2 W using ONNX policies.

  • Project
    pollen-robotics/microduck

    A control system for a tiny biped robot that uses reinforcement learning policies to walk, roll, and interact with its environment.

  • Project
    fanhao375/microduck-replica

    Microduck replica is a community project that reverse‑engineers the Pollen Robotics Microduck biped robot, providing editable CAD, reconstructed PCB schematics, BOMs, assembly videos, and a software fork with IMU‑I²C support and a web‑debug UI, enabling hobbyists and researchers to build and experiment with a reinforcement‑learning‑driven walking robot.

  • Dispatch
    Pollen Robotics Microduck: An Open-Source Biped for RL Training

    Pollen Robotics has launched Microduck, a 25cm open-source biped robot designed for reinforcement learning (RL) training via a sim-to-real pipeline using MuJoCo.

  • Project
    robotdoly/DOLY-DIY

    A complete open-source repository for building the DOLY robot, providing 3D printable hardware designs, electronic schematics, and a C++/Python SDK.