Project · Core Intelligence

Airspace Data & Intelligence API

Deliver airspace state without stripping away the evidence behind it.

Evidence levelE0Concept

From first observation to replay

One sequence, five moments. Step through them; nothing plays on its own.

Replay: Data table
ItemValue
Observed1, 4
Inferred2
Unknown3
Replay RD08Data status: SimulationEvidence level: E0 · ConceptQuestion answered: What was known at each moment?A replay line separates what was observed from what was inferred and what was unknown. Stepping forward or back reveals only what was known at that step.Limitation: A simulated sequence. It never shows predicted paths or knowledge gained afterwards.

Step 1 of 5Observations arrive

  1. Observations arrive
  2. They are correlated and fused
  3. The state is published with its evidence
  4. A stale source widens the band in the payload
  5. The earlier payload is requested and compared

Why this exists

Integrations tend to pass coordinates and drop confidence, provenance and unknowns. The receiving system then treats an uncertain reading as a fact, and afterwards no one can tell what was known at the time.

What it takes in

Each input adds something and leaves something unproven.

  • Governed state objectsThe state from the core platform, with its confidence, freshness, provenance and explicit unknown or stale fields.
  • Access and schema rulesWhich consumer may read which part of the state, under which version of the schema.
  • Time-based questionsRequests for what was known at an earlier moment, not only for what is known now.
Source presence: Data table
ItemValue
Source AContributes
Source BContext only
Source CNot yet available
Source presence RD03Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: Which sources contribute to this state?Three source markers connect to one state and carry a text status each. A source being present does not mean it is trusted; the label says what it contributes.Limitation: Sources are generic and their positions are not geographic.

How it works

Three steps, in the order the state is formed.

  1. Serialises each state with its confidence, freshness and provenance, and marks unknown or stale fields explicitly.
  2. Versions the schemas and applies authorization and audit to every consumer. No endpoint is published on this site.
  3. Answers questions of the form 'what was known at t', so a downstream system can reproduce a past picture.
  • A successful call is not the same as valid information.
  • Unknown is never serialised as zero or false.
  • The difference between known then and known now can be requested.

Where it sits in the architecture

The interface is the last link of the chain. It takes the unified state from the core platform, and from fusion where needed, and carries it to other systems with the evidence still attached, so that the receiving side can see how far to rely on each field.

Read the architecture

Data flow: Data table
StageValue
1Sources
2Model
3Policy
4Interface
5Consumer
6Audit
Data flow CH13Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: How does data flow through the architecture?A pipeline of six stages, from sources to audit, shows how data moves through the model, the policy and the interface to the consumer, with a record kept at the end.Limitation: A conceptual flow. It does not describe a deployed system.

Payload inspector

Interactive view

Data status: Illustrative

A generalized state object, invented for illustration. Change the input condition and read how each field responds; the marked lines are those that differ from the nominal payload.

Input condition
  1. {
  2. "state_id": "st-0417",
  3. "schema": "v3",
  4. "confidence": {
  5. "low": 0.72,
  6. "high": 0.86
  7. },
  8. "freshness_s": 2,
  9. "sources": [
  10. {
  11. "id": "src-a",
  12. "status": "live"
  13. },
  14. {
  15. "id": "src-b",
  16. "status": "live"
  17. },
  18. {
  19. "id": "src-c",
  20. "status": "live"
  21. }
  22. ],
  23. "altitude_band": "low",
  24. "unknown": [],
  25. "claim": "standard"
  26. }

How to read empty values

  • null Unknown. The value could not be supported; it is not zero and not false.
  • [] Nothing is recorded in this list. Absence is stated, not implied.
  • "softened" The claim wording was weakened because an input degraded.

Identifiers are invented. There is no endpoint, key or schema of a real interface on this page.

Evidence and assurance

What has been shown, to which level, and what has not.

Evidence ladder

  1. E0 · Concept reached
  2. E1 not reached
  3. E2 not reached
  4. E3 not reached
  5. E4 not reached
  6. E5 not reached

Project assurance panel

Mission and context
Let other systems use airspace state without stripping away the evidence that tells them how far to rely on it.
Primary contribution
A governed delivery of state objects whose confidence, freshness, provenance and unknowns travel with every field.
Decision value
Lets an integrating team see, in the data itself, which fields are firm and which are soft. The receiving system still decides how to use them.
Evidence status
Evidence level: E0 · Concept Concept and research direction only. No specification, interface or test result is claimed.
Next validation step
Draft the payload and null-semantics rules, then check them against synthetic degradation scenarios in simulation before any interface is built.
Behaviour when degraded
When a source goes stale, the affected fields carry a wider band and a softened claim, and fields that can no longer be supported become null. The payload stays well-formed.
Role in integration
Delivers states and history to approved consumers under versioned schemas, with authorization and audit. It exposes state; it does not accept commands for sensors.
Human review
An integrator reviews how each consumer uses the state, and a person reviews any action that depends on it.
Public-safe limitation
The payload shown is generalized and invented. It is not a real schema, endpoint or example of customer data.
Open unknowns
Which schema versions consumers will need, how long history must be kept and which authorization models sites will require are not yet known.

When inputs degrade

When a source goes stale, the affected fields carry a wider band and a softened claim, and fields that can no longer be supported become null. The payload stays well-formed.

  • Normal
  • Degraded
  • Unavailable
System health over time: Data table
Time (relative)Component 1Component 2Component 3
1NormalNormalNormal
2NormalNormalNormal
3NormalNormalNormal
4NormalNormalUnavailable
5NormalNormalUnavailable
6NormalDegradedNormal
7NormalDegradedNormal
8DegradedNormalNormal
9DegradedNormalDegraded
10NormalNormalNormal
11NormalNormalNormal
12NormalNormalNormal
System health over time CH07Data status: Synthetic dataEvidence level: E0 · ConceptQuestion answered: How has system health changed?Three bands, one for each of three example components, show normal, degraded and unavailable periods as different fills and patterns, so state never depends on colour alone.Limitation: Periods are invented and say nothing about any real component.

Where it applies

  • A site system that displays airspace state

    Why it matters here
    Its operators need to see which readings are soft.
    What it contributes
    Fields that carry band, freshness and provenance, so the display can show them.
    What it does not replace
    The site's own display design.
  • An archive that has to answer what was known then

    Why it matters here
    Reviews after an event depend on the picture as it was, not as it is now.
    What it contributes
    Time-based queries and a difference between two moments.
    What it does not replace
    The organisation's rules for handling and keeping evidence.

Role in the data flow

Delivers states and history to approved consumers under versioned schemas, with authorization and audit. It exposes state; it does not accept commands for sensors.

Validation path

Draft the payload and null-semantics rules, then check them against synthetic degradation scenarios in simulation before any interface is built.

Validation by stage: Data table
ItemDefinedPrototypeIntegratedField
Module ADoneIn progressNot startedNot started
Module BDoneNot startedNot startedNot started
Module CIn progressNot startedNot startedNot started
Validation by stage CH15Data status: IllustrativeEvidence level: E0 · ConceptQuestion answered: What is the validation status per stage?A grid of three example modules against four validation stages marks each cell as done, in progress or not started. Most cells are still open, as expected at this point.Limitation: Modules and statuses are placeholders and do not report real validation.

Limitations

  • An interface can only pass on what the state contains. If a source was absent, the payload says unknown; it does not fill the gap.
  • Consumers may drop evidence fields. The interface can make them available, but it cannot make a receiving system use them.
  • No interface is published and no schema is fixed. What is shown here illustrates the rules, not a product.
  • Delivery of state is not an assurance of its accuracy, and a well-formed payload says nothing about whether its content is right.

Related projects

Each link says why the two belong together.

Capabilities

Solutions

Technology