Pollen Robotics

A 25 cm open-source biped you train yourself with reinforcement learning. Playable out of the box.

Trained in sim, deployed on the real robot. This is the simulated twin the ducks were trained on.

Every behaviour is a policy you can retrain on your own machine .

Behaviours are learned in physics sim, on your machine or on Hugging Face Jobs.

Flat on its back to standing, all by itself, ready for the next command.

Every Microduck ships in one of four colourways . Same robot, same brains underneath - pick the shell that best fits you.

The real robot in real places - on desks, on the pitch, out at golden hour.

The robot is everything you need on day one . The packs add play gear and spare parts.

3x spare motors, 5x motor cables, 2x batteries, dual charger, 10x NFC tags, Hugging Face credit, screwdriver, screw pack.

Laser pointer, NFC polaroid, 2x rollers, ball, 10x NFC tags.

The SDK, the simulation and the full RL training stack are on GitHub. What the robot runs is what you can read, fork and retrain .

Builds on show, policies to swap , help when a leg does something strange. The community lives on Discord.