jrubiosainz/microduck-game-sprinter

MICRODUCK GAME sprinter

HannesVonEssen's microduck-running policy (Apache-2.0) fine-tuned with PPO on a CPU inside the Arena's own simulator: 1.05 s on the official seeds, 200/200 finishes and 0 falls on seeds 1-200. Ignores the command; simulation only.

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

What it is

A fine-tune of HannesVonEssen/microduck-running (revision 9f45af1f, policy.onnx sha256 007707dd7779…), HannesVonEssen's forward-running policy for the stock MicroDuck on flat ground (iteration 12,195, Apache-2.0), trained with Vottivott/microduck-playground@828d950 on pollen-robotics/microduck_rl@d424a0c. The network has the base's own shape (61 → 512 → 256 → 128 → 14, ELU) and keeps its input normalizer frozen; its weights were trained further with PPO on a CPU, inside the Arena's own runner and physics (MuJoCo 3.10.0, runner version 4), on this event's course:

• chain: HannesVonEssen's microduck-running policy (iteration 12,195) -> 1850 PPO iterations on sprint-2m (run hs1, whose first 25 iterations trained only a fresh critic); internal forward command 1.0 m/s
• selection: checkpoint it01850, picked among thirty-seven checkpoints of three sprint runs (hs1, s2 and s4), all run on the 200 seeds and noise rows of the table below: 1.048 s on the official seeds 1–5 in the trainer's check (after it also finished 20 held-out seeds, 1001–1020, with no fall), the lowest official time of the thirty-seven once the ones that fell are set aside (hs1 it2025 at 1.044 s, which fell on two of seeds 1–200; hs1 it1650 at 1.048 s, which fell on two of seeds 1–200; hs1 it1675 at 1.048 s, which fell on one of seeds 1–200), the lowest score on seeds 1–200 and no fall in any row;
• reward: forward progress while the race clock runs, small penalties for sideways speed, yaw rate, drifting toward the walls and falling; in run hs1 an episode also ends when the duck crosses the line, as the Arena's run does, with a one-off bonus equal to the reward of running at 1.5 m/s for every step left before the 2.5 s horizon; 96 parallel copies of the course in run hs1, 64 control steps each per iteration, the Arena's own start perturbation;
• command: a fixed internal command (forward speed 1.0 m/s, zero lateral speed, yaw rate, head and body) is written into the 13 command slots of the observation inside the graph, so the commanded twist is ignored.

No GPU, no new data, no paid compute.

Results

local run, Arena runner	score (non-finish = 50 s)	mean of finishers	finished
seeds 1–5 (the official set)	1.048 s	1.048 s (worst 1.08)	5/5
seeds 1-200	1.051 s	1.051 s (p90 1.08, worst 1.12)	200/200, 0 falls
seeds 1–5 with observation noise σ = 1e-04, 4 runs per seed	1.052–1.056 s (range)	—	20/20
seeds 1–5 with observation noise σ = 1e-03, 4 runs per seed	1.040–1.064 s (range)	—	20/20

The noise rows add Gaussian noise to every observation the policy sees: a stand-in for the float-rounding differences of another machine, which a chaotic gait amplifies.

arena score --event sprint-2m on this file (every speed of its sweep gives the same time, since the command is ignored; the sweep then reports the lowest):

sweep 21 commanded speeds 0.20–1.20 m/s
   0.20 m/s  1.06s
   0.25 m/s  1.06s
   0.30 m/s  1.06s
   0.35 m/s  1.06s
   0.40 m/s  1.06s
   0.45 m/s  1.06s
   0.50 m/s  1.06s
   0.55 m/s  1.06s
   0.60 m/s  1.06s
   0.65 m/s  1.06s
   0.70 m/s  1.06s
   0.75 m/s  1.06s
   0.80 m/s  1.06s
   0.85 m/s  1.06s
   0.90 m/s  1.06s
   0.95 m/s  1.06s
   1.00 m/s  1.06s
   1.05 m/s  1.06s
   1.10 m/s  1.06s
   1.15 m/s  1.06s
   1.20 m/s  1.06s

scored at 0.20 m/s
seed 1: finish   1.06s
seed 2: finish   1.08s
seed 3: finish   1.04s
seed 4: finish   1.02s
seed 5: finish   1.04s

score 1.048s · 5/5 finished · replay from seed 5

Local times are indicative: the Arena's machine can differ by a few hundredths of a second.

History

This repo's first revision (8f3bfe59) was Pollen's sprint-2m reference run unchanged inside a 37-parameter wrapper tuned with CMA-ES: 3.156 s on the Arena's board. Its second revision (f6241b55) was a PPO fine-tune of Pollen's sprint-2m reference (run s4's iteration-2075 checkpoint, after a chain of sprint and hill-climb runs): 1.084 s on the Arena's board. This revision replaces it with the PPO fine-tune of a different base described above.

Not for a real robot

This policy ignores the commanded twist: it always runs straight ahead at its own pace and never stops by itself. This fine-tune has only ever run in simulation and was trained without any sim-to-real randomisation; its base was trained with velocity pushes, centre-of-mass variation and initial tilt but, by its own card, is not validated on hardware either. Do not load it on a physical Microduck.

Files

• policy.onnx — one input obs [1, 61], one output actions [1, 14], 50 Hz, raw observations (the normalizer is baked in). Its metadata_props are the base's (without run_path) plus microduck_game_* provenance keys.
• manifest.json — schema 2 of the microduck policy manifest, built with Pollen's own microduck_rl publish code.
• LICENSE — the Apache License 2.0, under which the base is published and this fine-tune is distributed.
• NOTICE — where the weights come from and what this repo changed.

Credits

Base policy: HannesVonEssen/microduck-running by HannesVonEssen, under the Apache License 2.0, linked as base_model. Robot, simulator, Arena and publishing tools: Pollen Robotics. This fine-tune is a modified version of that policy, distributed under the same licence: see LICENSE and NOTICE. PPO fine-tune: the MICRODUCK GAME project (jrubiosainz), built with GitHub Copilot.

Resumen (español)

Participante del Sprint de 2 m de la Microduck Arena, solo para simulación. Es la política de carrera de HannesVonEssen (Apache-2.0) reentrenada con PPO en CPU dentro del simulador del propio Arena. Ignora la velocidad que se le ordena: no debe cargarse en un robot real.