Navigation

Navigation is the core competence of RMS-NAV: ingesting one image, predicting what the scene should look like from SPICE, and reporting the offset that brings the prediction into alignment with the data. This chapter is the top-level entry into every part of the navigation pipeline — from the per-image observation wrapper, to the predicted-scene NavModels, to the per-feature NavTechniques, to the orchestrator that runs them, to the per-image PNG overlay and the operator-curated regression library that anchors confidence calibration.

Each sub-chapter below stands on its own and links to the others. Start with the Navigation Overview for the architectural sketch and the per-image data flow, then read Class Hierarchy for the inheritance graph and the shared base classes the rest of the chapter assumes.

Architecture

• Navigation Overview
  • System architecture
  • Data flow
• Class Hierarchy
  • Top-level driver
  • NavBase
  • FeatureReliabilityGate
  • Dataset, Obs, and ObsSnapshot
  • NavModel
  • NavTechnique
  • NavFeature and NavFeatureGeometry
  • NavResult, ensemble, curator
  • Annotation

Pipeline subsystems

• Observations
  • Overview
  • ObsSnapshot
    • Backplane caching and thread safety
  • ObsInst
  • Per-instrument subclasses
    • Simulated-image instrument config: inherit / override / self-specify
  • Adding a new instrument
  • API reference
• Navigation Models
  • Stars
    • Star Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Simulated Star Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Bodies
    • Body Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Simulated Body Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Rings
    • Ring Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Simulated Ring Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Titan
    • Titan Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Simulated Titan Navigation Model
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
• Navigation Techniques
  • Shared infrastructure
    • DT Fitting (Shared Polyline-vs-Image Fitter)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Image Derivatives (Shared Gradient and Edge DT)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Confidence Calibration (Shared Sigmoid-of-Linear Combination)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Feasibility Reporting (Shared NavFeasibilityReport)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Per-Technique Diagnostics (Shared Dataclass Family)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Star techniques
    • Star Unique Match (StarUniqueMatchNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Star Field Pattern Match (StarFieldFromCatalogNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Star Refinement (StarRefineNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Body techniques
    • Body Limb Fit (BodyLimbNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Body Terminator Fit (BodyTerminatorNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Body Disc Correlate (BodyDiscCorrelateNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Body Blob Centroid (BodyBlobNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Ring techniques
    • Ring Edge Fit (RingEdgeNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
    • Ring Annulus Correlate (RingAnnulusNav)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
  • Titan techniques
  • Manual
    • Manual Navigation (NavTechniqueManual)
      • Overview
      • Theory
      • Configuration
      • Implementation
      • Examples
• Orchestrator Subsystem
  • Orchestrator (NavOrchestrator)
    • Overview
    • Theory
      • Pipeline stages
      • Exception-sandbox discipline
      • Glob-pattern model and technique filters
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
      • Call path
    • Examples
  • Per-Image State (NavContext)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Final Output (NavResult)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Per-Feature Post-Mortem (NavFeatureSummary)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Reproducibility Envelope (Provenance)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Image Classifier (NavImageClassifier)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Per-Instrument Settings (InstrumentSettings)
    • Overview
    • Theory
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • Ensemble Combine (ensemble + EnsembleConfig)
    • Overview
    • Theory
      • Step 1 — drop spurious
      • Step 2 — drop at-edge
      • Step 3 — single-link Mahalanobis grouping
      • Step 4 — pick the highest summed-confidence group
      • Step 5 — precision-weighted merge
      • Step 6 — disagreement and conflict penalties
      • Step 7 — confidence-rank assignment
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
  • JSON Curation (build_metadata_dict)
    • Overview
    • Theory
      • Float rounding policy
      • Allow-list discipline
      • Restrictions and assumptions
      • Sources of uncertainty
    • Configuration
    • Implementation
    • Examples
• Annotations
  • Overview
  • Class hierarchy
  • Pipeline
  • Configuration
  • Per-model annotation contributions
  • API reference
• Camera-rotation correction
  • Why per-instrument
  • The flag
  • Parameter vector
  • Per-technique strategy
  • Rank-deficient rotation
  • Ensemble combine in 3-D
  • JSON output
  • at_edge for rotation

Calibration and regression

• Image Library
  • Layout
  • Sidecar schema (schema_version 1)
  • The calibration process
    • Why the library is the right substrate
    • The calibration loop
    • Coverage taxonomy
    • Confidence-tier semantics
    • Per-instrument calibration anchors
  • Adding a new entry
  • Tolerance regimes
  • Regression baselines
    • What checks the baselines
    • How a baseline is created or updated