● Covered by 1 source · 1 reportLow impact1 positive
Pollen Robotics and Hugging Face launch Microduck, an open-source, retrainable robot
Key points
- Microduck is a small, retrainable robot.
- Behaviors are trained in physics simulation and deployed on the robot.
- The SDK, simulation, and RL training stack are open source.
- Seven pre-shipped policies are included and retrainable.
Introducing Microduck
Pollen Robotics, in partnership with Hugging Face, has introduced Microduck, a compact robot designed for hands-on reinforcement learning. The robot is presented as an accessible platform for developing and deploying new behaviors.
Retrainable Behaviors
A core feature of Microduck is its retrainability. Users can train new behaviors in a physics simulation environment, either on their local machine or using Hugging Face Jobs. These trained policies can then be directly deployed onto the physical Microduck robot. The process supports iterative refinement, allowing users to tune, retrain, and redeploy behaviors.
Open-Source Ecosystem
The entire software stack for Microduck is open source, licensed under Apache-2.0. This includes the Software Development Kit (SDK), the simulation environment, and the full reinforcement learning training stack, all available on GitHub. The robot runs on this open-source software, enabling users to read, fork, and modify its code. MuJoCo is used as the physics simulation engine for policy training.
Included Policies and Customization
Microduck ships with seven pre-programmed behaviors, or policies, which are also published and retrainable by the community. These include actions like walking, sitting, standing, kicking, grabbing, roller skating (when equipped), and self-righting. The robot is available in four different colorways, offering a choice in its physical appearance while maintaining the same underlying hardware and software.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Reporting from
Pollen Robotics, in collaboration with Hugging Face, has launched Microduck, a small, open-source robot designed for reinforcement learning. Users can retrain its behaviors in simulation and deploy them on the physical robot, with all software components available under an Apache-2.0 license.