Research snapshot August 6, 2026 Evidence current at 7e77fef

Sim2Claw / physical-to-simulation replay

Reality on the left. The simulator has to earn the right.

Sim2Claw turns physical SO-101 recordings into synchronized, evaluator-scored simulator evidence. The goal is not a plausible robot video. It is a replay whose camera, motion, contact, and consequences survive independent checks.

Held-out visual footage Same timeline No transfer claim

One held-out episode, frame for frame.

The physical C922 recording and frozen renderer-native candidate share a 5 fps evaluation timeline. No physical pixels are embedded in the simulated side.

This is the strongest honest visual result, not “nearly identical dynamics.” The evaluator passed its frozen numeric visual gates, then a human-readable sanity audit found that the board/background dominated the score while the robot and outside-board scene still differed materially.

0.844 Held-out full-frame similarity

Two episodes · 246 frames · visual metric only

4.26 mm Terminal center error

One outcome-informed object-dynamics episode · quarantined

0 / 6 Twin fidelity closure

No simulator promotion or physical transfer

Three real-to-sim wins worth showing.

Each is useful on its own. None is allowed to quietly inherit the authority of the next proof class.

Verified

Synchronized divergence footage

A 531-sample physical trajectory, simulator trace, and C922 recording can be viewed on one bound timeline. The resulting film makes the first contact and object-consequence mismatch inspectable instead of hiding it in an aggregate score.

Timeline
20 fps / 26.55 s
Actions changed
No
Bounded pass

Held-out renderer metric

A frozen camera, workcell, mesh renderer, and monotone response passed six of six numeric visual gates plus seven of seven integrity gates on two evaluator-held-out episodes.

Frames
246
Mean similarity
0.84445
Inspect visual closeout
Quarantined pass

Natural object-dynamics endpoint

With the robot driven through an unchanged 531-row measured-state trajectory, the simulated pawn evolved under natural dynamics and ended upright within 4.26 mm of the physical destination. The result reproduced without a refit, and all seven terminal outcome gates passed.

Final tilt
0.003864°
Other-piece motion
0.000041 mm
Inspect dynamics closeout

The footage is closer. The dynamics are not yet the same.

The best result remains useful precisely because the evaluator records where it fails.

Same-video semantic match Not achieved

Robot appearance, operator/cable structure, and the scene outside the board remain visibly different.

Contact and event parity Not achieved

No bilateral named-jaw contact; pawn motion and support loss arrive early; sample-260 tilt is 26.1°.

Cross-episode dynamics Not achieved

The 4.26 mm endpoint was selected on the same episode. It is not held-out calibration or predictive physics.

Physical transfer authority Closed

Twin fidelity remains 0/6 and strict task closure 0/11. Training, promotion, and transfer stay false.

A proof loop for embodied systems.

The problem

A simulation can look persuasive while disagreeing with the physical robot at the exact moment contact matters. Commands, measured motion, rendered pixels, object consequences, and physical authority are different claims.

The system

Sim2Claw binds SO-101 recordings, six-joint state and action streams, MuJoCo replay, camera registration, browser inspection tools, frozen splits, and independent evaluator receipts into one traceable workflow.

The contribution

The project makes failure legible. Exact action hashes prevent helpful edits; held-out evaluators prevent self-promotion; terminal negatives preserve the next experiment instead of being rewritten as demos.

  1. 01Record

    Physical video, timestamps, measured state, requested actions.

  2. 02Bind

    Hash episode identity, action tensors, runtime, and evaluator contract.

  3. 03Replay

    Keep actions fixed while isolating geometry, timing, and contact gaps.

  4. 04Evaluate

    Score frozen consequences on a separately owned CPU/fp32 path.

  5. 05Decide

    Promote, quarantine, or stop with a machine-readable receipt.

The result and its boundary travel together.

EvidenceObserved resultWhat it provesWhat it does not
OR26 · synchronized film 531 samples on one registered timeline Real/sim divergence can be inspected frame by frame Physics success or metric camera calibration
OR50 · dynamics endpoint 7/7 terminal gates; 4.26 mm center error A same-episode natural object-dynamics outcome can be reproduced exactly Event parity, held-out calibration, or transfer
OR91 · held-out visual 6/6 numeric and 7/7 integrity gates A frozen renderer clears its preregistered visual metric A semantic “same video” match
OR133C · stop boundary Residual identity unidentifiable from retained evidence The current corpus cannot resolve the remaining actor/cable residual A guessed fit or retrospective fidelity improvement

Next credible step: a fresh capture with independent actor/cable observability, metric camera inputs, and rigid jaw-bound landmarks—then a preregistered held-out dynamics comparison.

Inspect the current stop-boundary closeout