The evidence is proven or strong, but this capability still needs a capability-specific ESP32 or Rust runtime adapter.
Intended capability
Validate this target only after a capability-specific adapter and ground-truth study exist: Plant-wide tracking of humans and vehicles as distinct classes, generating near-miss heatmaps and live proximity warnings in aisles and yards without tagging anyone.
This describes the intended outcome. Readiness is needs edge adapter, evidence is class A, 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 Industrial scene
This is one recorded vertical scenario. It is not separate validation of every capability in the catalog.
01 The physics
Humans and forklifts have categorically different Doppler signatures (limb-swing micro-Doppler comb vs rigid-body monotone Doppler + mast vibration) — machine-vs-human discrimination is already proven. RTI provides simultaneous device-free tracks of both at ~0.5-1m, sufficient for zone-level proximity logic.
02 Shared processing path
Micro-Doppler class features (stage 2, proven discriminator) -> RTI multi-target differential tracking (stage 4) -> anomaly manifold on the joint human-vehicle trajectory distribution (stage 3) to surface recurring near-miss hotspots rather than single alarms.
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
Machine-vs-human already class A; multi-target tracking validated on WiMANS and OPERAnet; own-hardware trial in one live warehouse aisle with UWB ground-truth tags on one forklift and two workers for a week.
04 Commercial hypothesis
A near-miss heatmap of the whole warehouse with zero tags and zero cameras — sold to the safety manager who can't get camera analytics past the works council.