B1 · Proof · Robot cell Sensor simulated
The safety zone is not a drawing.
It is a calculation.
The minimum distance comes from response time, approach speed and sensor geometry. The cell computes it on video — for this run without S_M, because the arm is at rest.
Prove the safety zone — without sampling.
Safety-zone evidence today is manual work: dimensions from drawings, spot checks at the cell, a report written afterwards. Meanwhile the trend moves toward fenceless, collaborative cells: without a guard fence, sensing and SSM carry the safety function alone — and every change to the cell needs the evidence again.
Measure the scene, calculate S_min for this run, run the SIL test in the twin — the protocol writes itself while the run is in progress.
The arm stops before the zone. Protocol #0117 names the inputs, the formula without S_M, and 532 mm.
Reproducible instead of sampled: same setup, same result. Collaborative cells per ISO/TS 15066 treat SSM as one of four methods. This run computes without S_M; the check against the standard edition is still open.
The protocol writes itself during the SIL test.
No report written afterwards. While the test runner executes the script, the protocol writes itself — every step timestamped: scan, geometry, S_min, sensor status, report. Time-synchronized with the recording, restarted on every loop.
▸ d = 532 mm → zone holds · sensor simulated
532 millimeters — computed, not guessed.
The controls sit on this run's inputs. Formula without S_M, because the arm is at rest. +S_M applies only when the hazard point moves — the switch in the calculator.
This run: without S_M, the arm is at rest. +S_M appears only when the hazard point moves.
Where the numbers come from.
No value stands alone: the graph shows which standard feeds which parameter into 13855 — and what changed with the 2010 → 2024 edition. This run uses S = (K×T)+DDS+Z. +S_M is the variant for a moving hazard point and is not included here.
What changed — and what didn't.
The same cell — with the arm in motion.
The video shows the SIL test with the arm at rest, so it computes without S_M. The SIL simulation moves the arm. +S_M is the variant for a moving hazard point; this run does not apply it. In the calculator you switch it on separately.
Open the SIL simulation →On record, not just claimed.
This run's calculation is in the evidence block above. The files below are the recording and the SIL run. They do not confirm the standard edition.
Every measurement becomes a node.
Reads
3D-ToF point cloud · distances to the zone boundary · from recording (sensor sim)
Transforms
Zone node + S_min edge · S=(K×T)+DDS+Z, this run without S_M
Writes back
Sealed protocol #0117 · measurement bar · result 532 mm
Honest status.
What runs, what is simulated, what is still unproven — as of 2026-07-30.