Reference

Default Intent Axis Definitions

Version: 0.2
Status: Draft
Lineage: Default axis set used by M45

Normative framing

  • Axes define how intent is structured, not what intent must exist
  • Axes are independent and optional
  • Absence of an axis does not imply missing intent
  • Presence of an axis does not imply completeness

Intent elements in intent snapshots are classified under these axes via the axisId field. The axes enabled for a project are declared in the Project Intent Profile.

If project-specific axes are defined, they should follow the same structure and clarity as the defaults below. See the Axis Definition Schema for the formal structure.


Default axis set

The following axes are included by default in M45.

Each axis definition includes:

  • A semantic definition
  • Inclusion criteria
  • Typical artifact sources
  • Examples and non-examples (each with a short rationale explaining why it is an example or non-example)

Axis: component

Definition

A component is an identifiable physical or logical system part treated as a distinct unit in engineering artifacts.

Inclusion criteria

  • Has a recognizable boundary in artifacts
  • Is referenced consistently enough to be disambiguated
  • Can participate in relationships with other intent elements (via the intent graph)

Typical sources

  • Requirements naming system parts
  • Architecture descriptions
  • Interface documentation
  • Design diagrams

Examples

textrationale
Flight Management SystemTreated as a bounded unit in requirements/architecture artifacts.
Navigation Software ModuleA named logical subsystem with a stable boundary in design documentation.
GNSS ReceiverA distinct hardware element referenced consistently across artifacts.
Power Distribution UnitA physical unit with clear interfaces and responsibilities.

Non-examples

textrationale
"shall compute"Describes an action; it belongs under the function axis.
"high availability"A quality/constraint theme, not a bounded system part.

Axis: function

Definition

A function is an intended transformation, computation, or action performed by the system or its components.

Inclusion criteria

  • Describes what is done, not how it is implemented
  • Can be scoped by modes or scenarios
  • Can be traced to one or more artifacts

Typical sources

  • Functional requirements
  • Operational descriptions
  • Interface behavior descriptions
  • Test intent and procedures

Examples

textrationale
Compute navigation solutionAn intended computation the system performs.
Update navigation outputsAn intended transformation of internal state into outputs.
Monitor signal integrityAn intended monitoring action performed by the system.
Generate alertAn intended system action that produces a notification.

Non-examples

textrationale
"shall be reliable"A quality expectation/constraint theme, not a specific transformation/action.
"shall comply with standard X"A compliance/assurance claim, not a system function.

Axis: mode

Definition

A mode is a persistent operational state that alters system behavior or interpretation of inputs and outputs.

Inclusion criteria

  • Meaningfully distinct from other states
  • Persists until exit conditions apply
  • Affects functional behavior or constraints

Typical sources

  • Requirements referencing operational states
  • Mode logic descriptions
  • State machine artifacts
  • Degraded-operation descriptions

Examples

textrationale
Normal operationA persistent operational state with distinct behavior.
Initialization modeA persistent state with different interpretation of inputs/outputs.
Protected modeA persistent state that constrains/changes behavior.

Non-examples

textrationale
One-time eventsEvents do not persist as modes.
Failure descriptions without state semanticsA fault description without an operational state definition.

Axis: scenario

Definition

A scenario is a contextual situation or set of conditions under which specific behavior is intended to apply.

Inclusion criteria

  • Describes circumstances, not system state
  • Can scope behavior, constraints, or expectations
  • May reference modes but is not equivalent to them

Typical sources

  • Operational concepts
  • Degraded condition descriptions
  • Use-case narratives
  • Scenario catalogs and operational narratives
  • Test procedures and test case catalogs

Examples

textrationale
Loss of GNSS signalA contextual condition that scopes behavior/constraints.
Transition from ground to flightAn operational context under which behavior changes.
Degraded sensor availabilityContextual conditions that scope expected behavior.
GNSS signal loss during approach with inertial fallback enabledA concrete scenario used to exercise and observe hazard-relevant behavior (often used in V&V).
Degraded sensor availability with alerting thresholds at nominal limitsA concrete scenario used to validate alerting behavior under defined conditions.

Non-examples

textrationale
Persistent operational statesThese are modes, not scenarios.
Abstract conditions without behavioral impactWithout scoping behavior/constraints, it is not a meaningful scenario axis element.

Axis: interface

Definition

An interface is an identifiable interaction boundary between system elements or between the system and its environment.

Inclusion criteria

  • Has identifiable endpoints
  • Supports one or more interactions
  • Directionality is known or explicitly undefined

Typical sources

  • Interface control documents
  • Architecture diagrams
  • Integration test descriptions

Examples

textrationale
FMS to Autopilot interfaceA clear interaction boundary between two system elements.
External data bus interfaceA defined boundary supporting interactions and signals.

Non-examples

textrationale
Vague information flow without endpointsLacks identifiable endpoints/boundary.

Axis: signal

Definition

A signal is an identifiable unit of information exchanged over an interface.

Inclusion criteria

  • Has a recognizable name or description
  • Has an implied producer and consumer
  • Can be traced to an interface or interaction

Typical sources

  • Interface specifications
  • Data dictionaries
  • Telemetry definitions

Examples

textrationale
Position estimate outputA named information item exchanged across an interface.
Health status flagA distinct data item with producer/consumer semantics.
Control input commandA discrete information unit sent across an interface boundary.

Non-examples

textrationale
Abstract "system data" without defined meaningNot a distinct signal with clear semantics.

Axis: hazard

Definition

A hazard is a system-level hazardous state that could lead to unacceptable loss (not a component failure mode).

Inclusion criteria

  • Describes an unsafe system state with loss consequences
  • Is stated at system level (may involve multiple components)
  • Is meaningful independent of a specific root-cause failure

Typical sources

  • Safety analyses (STPA/FTA/FHA)
  • Hazard logs
  • Requirements and operational descriptions

Examples

textrationale
Loss of navigation capability during approachAn unsafe system condition with safety consequences.
Erroneous altitude guidance displayed to crewA hazardous system state that can lead to unsafe decisions.

Non-examples

textrationale
GNSS receiver power supply failsA component failure mode; it may contribute to a hazard but is not itself the hazardous state.
Loss of GNSS signalA context/scenario; the hazard is the unsafe system state that results if uncontrolled.

Axis: safetyConstraint

Definition

A safety constraint is a constraint derived from hazards that limits behavior or design to prevent or mitigate unsafe system states.

Inclusion criteria

  • Restricts behavior or design in a hazard-relevant way
  • Can be traced to one or more hazards it mitigates
  • Is actionable enough to be reviewed for adequacy/coverage

Typical sources

  • Safety requirements
  • Safety analyses
  • Requirements and verification plans

Examples

textrationale
The system shall transition to inertial navigation within 100 ms of GNSS signal loss during approachConstrains behavior to mitigate loss of navigation capability.
The system shall inhibit autopilot engagement when air data validity is unknownConstrains control to mitigate hazardous states.

Non-examples

textrationale
The system shall be safeToo vague to be an actionable constraint.
The system shall compute navigation solutionA function statement, not a hazard-derived constraint.

Axis: uca

Definition

An unsafe control action (UCA) is a control action that, under certain contexts, can lead to hazards or losses (STPA-style).

Inclusion criteria

  • Identifies a control action
  • Includes context that makes it unsafe (provided/omitted/too early/too late/too long)
  • Can be connected to hazards via reasoning/artifacts

Typical sources

  • STPA analyses
  • Safety analyses and review notes

Examples

textrationale
Autopilot engages climb mode when airspeed is below minimum safe thresholdA control action unsafe in a specific context.
Navigation system suppresses integrity alert during approachAn omission that can be unsafe in context.

Non-examples

textrationale
The autopilot shall have a climb modeA capability/feature statement, not an unsafe control action.
Loss of navigation capability during approachA hazard, not a control action.

Axis: verificationObjective

Definition

A verification objective captures verification/validation intent: what must be verified or validated, and (when applicable) the criteria used to judge acceptability.

Inclusion criteria

  • States what is to be verified/validated (not the evidence itself)
  • Can be linked to intent elements and to planned methods
  • Is reviewable for completeness and adequacy
  • Includes criteria (thresholds, timing bounds, pass/fail conditions) when those criteria are part of the objective's meaning

Typical sources

  • Verification plans
  • Test plans and verification matrices
  • Certification planning artifacts
  • Acceptance criteria documents
  • Test specifications

Examples

textrationale
Verify navigation accuracy constraint is satisfied under approach conditionsStates what must be verified and under what context.
Validate integrity alert behavior with operational flight test dataStates a validation objective and expected evidence class.
Horizontal position error shall remain <= 10 m (95%) during nominal approachA checkable acceptance criterion used to judge whether the navigation accuracy objective is met.
Alert shall be issued within 2 seconds of integrity threshold crossingA measurable pass/fail criterion that makes the objective reviewable and testable.

Non-examples

textrationale
Test report TR-123 shows passAn evidence result, not an objective.
Compute navigation solutionA function; it is what the system does, not what must be verified.
High performanceToo vague to serve as an objective or as an acceptance criterion.