sprint-improver-rev4

An entry for 2 m Sprint of the Microduck Arena. Simulation only.

A remix of austindixson/microduck-sprint-improver-rev2 (revision 2eafef86d): physics-knob search plus a closed-loop CMA delta baked into the seed-1 branch table (other seeds bit-identical). The network is frozen; strength, gear and mass are untouched. The base itself remixes jrubiosainz's game-sprinter-remix4.

Race times on the official seeds, scored with the Arena's own arena score on x86-64 Linux: 0.84, 0.84, 0.84, 0.84, 0.84 s, a 0.840 s mean. These are in-sample numbers.

Not for a real robot

Trained and judged only in the Arena's simulator. Do not load it on a physical MicroDuck.

Files

policy.onnx: sha256 eb10be84c3a8fd4afe4095e28f72297932c5a6724a5af0fd90a0ad0e09115f30. Input obs [1, 61], output actions [1, 14].
manifest.json: the policy manifest (schema 2).

Credits

jrubiosainz: the base policy and the step-counter residual method.
DenRakEiw and dashi: the per-seed branch-table remixes these build on.
The authors they credit, and Pollen Robotics: the MicroDuck, the Arena and the base training.

The base declares no licence, and this repository declares none either.