Regulatory Framework
Precision GNSS in the United States operates inside a layered federal framework. The NOAA National Geodetic Survey operates the CORS network, Continuously Operating Reference Stations, which is the public reference infrastructure for centimeter-grade positioning across CONUS, Alaska, Hawaii, and the U.S. Caribbean. CORS feeds the National Spatial Reference System, the legal datum framework for surveying, mapping, and federal geospatial work. The NSRS-2022 modernization replaces NAD 83 and NAVD 88 with new geometric and geopotential reference frames, tied through CORS into the International Terrestrial Reference Frame via IGS contributions.
On the augmentation side, the FAA Wide Area Augmentation System (WAAS) provides space-based augmentation for aviation use; FCC Part 15 governs unlicensed RTK base-station emissions in the 902 to 928 MHz and other bands where private RTK bases operate; NGS OPUS (Online Positioning User Service) provides public post-processing of static GNSS observations against CORS. The IIJA's Geo-Rail provisions extend federal precision-positioning infrastructure expectations into rail corridors. Commercial network-RTK and virtual-reference-station (VRS) services operate as licensed overlays on top of or in parallel with CORS, providing the correction service that makes single-base RTK obsolete in the geographies where those operators have built out coverage.
Architectural Requirement
The architectural requirement for modern precision positioning is centimeter-grade horizontal accuracy, sub-decimeter vertical accuracy, traceable lineage to the NSRS, integrity monitoring, and availability across the full operating geography. Surveying applications need legal-grade datum traceability. Agricultural applications need field-level precision across the entire operating area. Autonomous-vehicle applications need lane-level precision plus integrity. Mining applications need pit-floor precision plus operating availability. Construction applications need machine-control precision plus continuity across the build site. Defense and expeditionary applications need precision in geographies where CORS does not exist and where commercial RTK services do not operate.
The requirement that no current architecture satisfies is geographic universality. CORS density is excellent in populated CONUS and thin in the Mountain West, the Great Plains, Alaska's interior, and most of the Pacific. Commercial RTK service coverage tracks population density and customer concentration; it is excellent in the Corn Belt and absent across most of the Mountain West and the desert Southwest. Private RTK bases extend coverage but do not produce traceable NSRS lineage. The architectural requirement is therefore a primitive that produces NSRS-traceable centimeter-grade positioning in geographies where the centralized economics do not close, while composing with CORS where CORS exists.
Why Procedural Compliance Fails
Procedural compliance with the centralized model fails the universal-geography requirement in three structural ways. First, the reference-station capital and maintenance economics do not close in low-density geographies. NOAA cannot site CORS where the federal-use justification is thin. Commercial RTK operators cannot site VRS networks where the subscriber count does not justify the buildout. Private base operators cover their own operating site but produce nothing for the broader user community. The cumulative result is a coverage map that is excellent where the economics close and absent everywhere else, and the gap is structural rather than incidental.
Second, the NSRS-traceability requirement amplifies the gap. A private RTK base produces centimeter-grade relative positioning but does not, on its own, produce NSRS-traceable absolute positioning. OPUS post-processing requires CORS observability of the rover, which is precisely what is missing in the geographies the rover most needs to operate in. The user can have precision or traceability but not both, and the legal use cases (cadastral survey, federal mapping, regulated construction) require both.
Third, the centralized model has no path to scale into the geographies where the demand is now growing fastest. Autonomous off-highway equipment, agricultural autonomy, mining autonomy, defense expeditionary autonomy, and rail-corridor monitoring under IIJA Geo-Rail are all expanding into the geographies that CORS and commercial RTK do not cover. Procedural compliance with the existing model produces no architectural primitive for closing the gap; it produces only the recommendation that more reference stations should be built, which the economics have already declined to do.
What the Capability-Awareness Layer Provides
The capability-awareness layer disclosed in United States Patent Application 19/647,395, combined with the AQ portfolio's distributed mesh-coordinates substrate, provides a fleet-emergent alternative to centralized reference infrastructure. The substrate is the data layer: operating units passing fixed references, survey monuments, infrastructure features, persistent natural features, contribute credentialed observations that accumulate into a shared position estimate for each reference, refined as fleet contributions accumulate. Capability Awareness is the runtime layer that makes that substrate operationally trustworthy, and it is the disclosed inventive contribution this application rests on.
Under the disclosed primitive, each rover advertises a capability envelope, a structured, living description of its present positioning affordances across defined dimensions, including model access (the correction sources and reference observations it can consume), locality (its geographic and network position relative to those sources), execution guarantees (the integrity and availability characteristics of its solution), and sensor interfaces (its GNSS, inertial, and ranging instruments). Resolving a position is treated as a capability-native computation: rather than a binary fix-or-no-fix gate, the system performs a per-dimension match in which each dimension resolves to satisfied, unsatisfied, or conditionally satisfiable, and the aggregate determination is one of four bounded outcomes. A position is structurally possible when the required precision class can be produced now from available credentialed contributors; structurally impossible when no contributor set can supply it; deferred when temporal executability forecasting projects, within a confidence-bounded window, that the necessary contributors will become available as the fleet moves or more observations accumulate; or rerouted when the precision can be produced from a different contributor set or a different substrate. Every determination is recorded as a structured, auditable record carrying the contributing observation set, its credential chain, and the propagated uncertainty bounds.
This is the property that compliance with the centralized model cannot produce in off-coverage geographies: a defensible, auditable answer to "how good is this position, on what evidence, and when it will get better." The temporal forecast distinguishes a position that will sharpen as the fleet accumulates passes (deferred) from one that will never reach the required class on the current substrate (impossible), which prevents a rover from waiting indefinitely on precision that is not coming. Capability genealogy keeps an append-only, integrity-protected history of how each reference's capability envelope evolved, giving forensic traceability for any disputed fix. Where CORS observability is available, the layer consumes CORS as one high-credential contributor among many, and the resulting estimate inherits NSRS-traceability through that contributor's credential chain. Where CORS is absent, the layer operates on the fleet-emergent substrate and produces an estimate whose traceability runs through the credentialed reference lineage rather than through a single reference station.
The economics differ structurally from the centralized model. Maintenance burden distributes across the operating fleet rather than concentrating in dedicated reference operators. Geographic coverage is a function of fleet density rather than of capital-siting decisions. The layer composes with NSRS-2022 by treating NSRS-anchored references as high-credential contributors whose observations dominate the determination where they are available. Through envelope negotiation, a rover whose capability falls short for a required precision class can advertise modifications, engaging an additional correction channel or a dormant ranging sensor, and have the determination re-evaluated against the cost of that change versus rerouting. The architecture is additive to CORS, not adversarial to it.
Compliance Mapping
NSRS and NSRS-2022 datum traceability is preserved through the credential chain on contributing observations: any reference whose lineage traces to a CORS-anchored survey carries that lineage forward into every determination it participates in, and the propagated uncertainty bounds make the resulting precision class explicit. NGS OPUS post-processing is composed with rather than discarded: where a rover's capability envelope can reach CORS observability, OPUS-anchored references enter as high-credential contributors; where it cannot, the substrate's continuous refinement provides the off-coverage analogue, with the capability determination recording exactly which contributor set produced each estimate. IGS global reference-frame consistency is maintained through the credential chain into ITRF-anchored references where they are present.
FAA WAAS augmentation continues to operate within its medium and contributes as a credentialed observation channel where it improves the estimate. FCC Part 15 governance of private RTK base emissions remains entirely unaffected; the capability-awareness layer does not require base-station emissions to operate, although it composes with them as additional credentialed contributors when they are present, with each rover's capability envelope recording whether such a channel is currently within reach. IIJA Geo-Rail expectations for rail-corridor precision-positioning infrastructure are satisfied by the substrate's ability to produce NSRS-traceable precision along corridors that the centralized model under-serves, with the credential chain providing the federal-grade audit basis the IIJA presumes.
Adoption Pathway
Adoption begins inside a single high-density operating geography, a mining pit, an agricultural operation, a construction site, a rail corridor segment, where the operating fleet is large enough that the consensus refinement converges quickly and the marker memory becomes operationally useful within the first season. The substrate operates alongside whatever CORS or commercial RTK is already in use; the rovers consume CORS where available and contribute their own observations into marker memory continuously. The first measurable adoption gain is reduced dependency on private RTK base maintenance and reduced exposure to commercial RTK service outages.
Within a single operating geography, the primitive also resolves the recurring private-base maintenance problem. Mining and large agricultural operations that today maintain their own RTK base stations carry the burden of base-station siting, antenna stability monitoring, ionospheric correction validation, and operator training; each of those costs is real and recurring. The fleet-emergent substrate consumes the existing private base as one credentialed contributor rather than depending on it as a single point of failure, which means the operator can let private bases age out of service without losing precision continuity. The substrate also resolves the multi-vendor RTK interoperability problem that operators with mixed-vendor fleets currently navigate through per-vendor correction-stream subscriptions; every contributing observation enters the substrate through its credential chain and the consensus refinement is vendor-agnostic by construction.
The second adoption stage is geographic extension. A fleet operating across multiple sites, or multiple fleets operating in adjacent geographies, share marker memory through the credentialed substrate. The shared memory extends precision coverage into the inter-site geographies where neither operator individually justifies private infrastructure. Cross-domain composition further compounds the effect: agricultural autonomy fleets operating in a Mountain West basin, long-haul autonomous trucks along the same corridor, rail-corridor inspection platforms along an adjacent right-of-way, and mining haul trucks in a basin pit all contribute marker observations into the same credentialed substrate. Marker memory becomes denser with each participating fleet, consensus precision improves for every participant, and the operating geography that no single fleet could justify infrastructure for becomes the geography that all participating fleets share. The federation property is structural rather than negotiated: each fleet's credential chain is preserved, each fleet's contribution is auditable, and each fleet benefits from the others' contributions without surrendering operational sovereignty. The third stage is federal and standards integration: NOAA NGS, the FAA, the FCC, and IIJA Geo-Rail program offices consume the credentialed substrate as a federally-recognized contributor to NSRS-2022 maintenance and integrity, and the substrate becomes one of the mechanisms by which the NSRS itself is densified into the geographies that the centralized siting model has not reached. Across all three stages, the disclosed capability-awareness layer is what makes the substrate usable as legal-grade infrastructure rather than a best-effort correction pool: every position carries a bounded, audited determination of how good it is and on what evidence. That is the architectural layer where centralized RTK reference networks meet their structural geographic limit and where decentralized, credentialed, fleet-emergent precision positioning takes over.
Embodiments and Deployment Options
The disclosed primitive admits several embodiments across the precision-positioning domain. By compute and instrument class, a rover embodiment ranges from a survey-grade multi-frequency GNSS receiver with an inertial unit and LiDAR, to a low-cost single-frequency agricultural guidance unit, to a rail-inspection platform with odometry and fixed-track references; in each case the capability envelope describes the actual instruments present and the determination is computed against the precision class the task requires. By contributor mix, a deployment may run CORS-anchored (the envelope reaches CORS observability and the determination resolves possible against NSRS-traceable contributors), private-base-augmented (a local RTK base is one contributor among many), or fully fleet-emergent (no fixed reference station is reachable and the determination rests on the accumulated credentialed substrate). By domain, embodiments include cadastral and federal survey requiring legal datum traceability, agricultural and off-highway autonomy requiring field-level precision across an operating area, mining requiring pit-floor precision with operating availability, construction machine control requiring continuity across a build site, and defense and expeditionary operation requiring precision where neither CORS nor commercial service exists. By federation scope, a deployment may be single-fleet single-site, single-fleet multi-site with shared substrate, or multi-fleet cross-domain federation in which each participant's credential chain is preserved and each benefits from the others' contributions without surrendering operational sovereignty. By outcome handling, a deployment may configure how deferred and rerouted determinations are surfaced to the operator, for example holding an autonomous task until a confidence-bounded window opens, rerouting to an alternative contributor set, or degrading gracefully to a lower precision class with the reduced class explicitly recorded.
Disclosure Scope
The capability-awareness mechanisms this application relies on, the per-dimension capability envelope, capability-native computation with three-valued matching and the four bounded outcomes (structurally possible, impossible, deferred, or rerouted), temporal executability forecasting with confidence-bounded windows, uncertainty propagation, envelope negotiation, and capability genealogy, are disclosed in United States Patent Application 19/647,395. This article is a public, dated, enabling description of applying those mechanisms to GNSS precision positioning; it is not itself a claim of those mechanisms and does not disclose capability metrics or benchmark figures beyond what the cited application supports. The distributed mesh-coordinates substrate referenced as the data layer is the subject of separate AQ portfolio filings.
