All capabilities
Cobot Cell Witness Shared-family mapped

Still Hold

Still Presence Hold: The Motionless Human Inside the Cell

Keeps track of a motionless person inside the robot cell — the crouched maintenance worker no log ever shows.

Feasibility path · not a production offer

Shared Cell Witness Kit

This offer is scoped to this capability only. It does not promote sibling catalog entries or replace physical deployment validation.

~$350 parts · 6 Nodes · ~150 minReview feasibility scope →No charge today · configuration and quote confirmed first
Readiness Shared-family mapped

Mapped to a shared offline runtime family and usable with a recording or compatible ESP32 CSI stream.

Evidence Strong theory Supported hypothesis

Published physics or adjacent results support the hypothesis; capability-specific product and site validation are still required.

Runtime family Presence

Calibrated motion and presence score from CSI variance and change statistics.

When no ESP32 is connected

No compatible ESP32 stream is connected. Use the bundled Cobot recording; it demonstrates the shared Presence runtime, not independent proof of this capability.

Intended capability

Pilot this intended outcome through the shared Presence family, then validate it against site-specific ground truth: Holds the presence of a motionless human inside the cell's human-access volume — the maintenance worker crouched at the base, the programmer standing at the pendant — for as long as they are there, and writes the interval against the cell's observed motion state. US robot-fatality studies covering 1992-2017 put most deaths near stationary machines during maintenance, with the victim's presence in no machine log; this row is that missing record, kept by the space itself. An evidence layer, never a safety function: it records and never intervenes, the certified stack remains authoritative, and ISO 13849 is out of scope.

This describes the intended outcome. Readiness is shared-family mapped, evidence is class B, and a catalog mapping or recording is not proof of this outcome at a real site.

Solution blueprint

See the environment before installing it.

This exact kit is one of 191 first-class designs. It includes geometry, objects, nodes, wording, scenarios, installation, limitations, and catalog-bound readiness.

Shared-family mapped

No rendered revision is available yet.

Bundled recording

A related Cobot scene

This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.

01 The physics

CSI is honest about its blindness here: a still human writes nothing into a 6.6 Hz motion field. The LD2410C cone aimed into the human-access volume holds them anyway — mm-scale chest displacement at 24 GHz (lambda ~12.5 mm) keeps a breathing, motionless person present to ~6 m at millisecond refresh. The hold is scoped to the cone: an arm or shuttle transiting it dominates the single composite target and masks the person, so hold intervals are gated to machine-quiet windows or to cones aimed where the machine never sweeps.

02 Shared processing path

LD2410 static-target energy in the in-cell cone -> entry handoff from cobot-threshold-truth (a crossing opens a candidate presence) -> static-hold state machine: presence persists until an exit crossing is seen, never released on signal fade alone -> machine-transit gating (CSI arm-cycle energy marks mask windows as no-assessment) -> presence interval with motion-state overlay -> signed row to the session ledger and certificate.

This is a capability design path. Components may be shared with other catalog entries; it is not presented as a unique algorithm.

03 Validation plan

Own-cell protocol: 10-minute motionless holds (standing, crouched, prone) at ranges 1-6 m and off-axis angles, 50+ trials; hold-ratio and time-to-loss curves published per posture; deliberate arm-proxy sweeps through the cone to measure mask onset and recovery; a multi-day empty-cell false-hold soak with powered idle machinery — servo hold dither, fans, cable sway, HVAC cycling — with the false-hold rate published as the row's central metric; exit-only decrement logic verified line-by-line against video ground truth.

04 Commercial hypothesis

'Still Hold' — for the EHS investigator and the plant owner: the only record that a person was inside the cell at 02:13, kept by the space itself, because no fleet log holds a human who did not badge in.