Mapped to a shared offline runtime family and usable with a recording or compatible ESP32 CSI stream.
Intended capability
Pilot this intended outcome through the shared Zone dwell family, then validate it against site-specific ground truth: Compiles an append-only ledger of human-robot co-presence events — a human verdict and a machine-motion verdict in the same zone over the same interval — each carrying its sensor evidence: the CSI band verdicts, the cone's range-gated hold where covered, and an event-bounded person-confirm snapshot where a camera node watches the zone entry. Co-presence is the precondition of any separation question, and it is the whole claim: the certified stack owns separation live, and this ledger never attests whether separation held — only how often a person and a moving robot actually shared a zone, over months, at zone granularity. It is frequently the only evidence a human was in an autonomous zone at all, because fleet managers track robots, not people. Evidence, never enforcement, and never a distance measurement.
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.
No rendered revision is available yet.
Bundled recording
A related Fleet scene
This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.
01 The physics
Human and machine verdicts come from envelope statistics on the full-rate CSI path: gait-band energy at 1.5-2.5 Hz with cadence jitter marks the human; gait-band absence plus sustained smooth motion energy marks the translating machine — a driving AMR writes no sub-Nyquist periodicity, so periodicity claims are reserved for cyclic machinery and repeated loop traffic, where sub-Nyquist periods actually exist. The cone's breathing hold carries the still human CSI loses; the camera's on-device person-confirm attaches pixel evidence to the instant of zone entry and can structurally never stream. One link's CSI is the superposition of every mover in its Fresnel volume, so a walker beside an AMR yields a composite envelope — the ledger degrades that case to 'human present among machines' and says so.
02 Shared processing path
Full-rate CSI human/machine separation per zone (gait band for the human; gait-band absence + sustained smooth motion energy for translating robots; periodicity only for cyclic sources) -> cone static-hold join for still-human coverage -> event-bounded camera person-confirm at instrumented entries -> co-presence event assembly (human verdict + machine-motion verdict, same zone, overlapping interval) -> integrity-chained append to the ledger -> periodic roll-up into the certificate stream with per-event class tags.
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-hardware protocol: choreographed co-presence sessions — walker plus AMR in shared and adjacent zones, still human in cone, cart-pusher confounders; score event recall, composite-envelope degradation handling, and false co-presence rate over a two-week ambient soak.
04 Commercial hypothesis
'Co-Presence Ledger' — for the site EHS and compliance owner whose fleet is built under ISO 3691-4 and whose user-side duties ANSI/A3 R15.08-3 anticipates: a months-long independent record of every time a person and a moving robot shared a zone, with evidence attached — the record the fleet log structurally cannot hold.