core analysis module

Deviation Engine

Detect, quantify, and contextualize execution drift as it happens. This module identifies when movement departs from intent, measures the magnitude and duration of that divergence, and preserves defensible evidence for instruction, review, and readiness assessment.

KEY SIGNALS SURFACED
  • Off-route deviation onset
  • Deviation magnitude and direction
  • Duration of sustained deviation
  • Partial recovery vs full recovery
  • Recurring deviation patterns across sessions
  • Intent-execution divergence zones
readiness intelligence

Why This Signal Matters

Most mission failures do not begin with a single dramatic error. They begin with small, explainable deviations that compound over time. Without a dedicated mechanism to detect and quantify drift, teams are forced to reconstruct failure after the fact instead of correcting behavior while it is still recoverable. The Deviation Engine makes deviation visible early, objectively, and in context.

What Becomes Visible

This module transforms raw movement telemetry into actionable evidence by isolating when deviation begins, how long it persists, and whether recovery occurs.
Execution Signals Exposed
  • Exact point where deviation first occurred
  • Degree of divergence from expected behavior
  • Whether deviation was transient or sustained
  • Patterns of repeated deviation across sessions
Decisions This Enables
  • Earlier instruction intervention
  • Objective after-action review
  • Reduced narrative reconstruction
  • Pattern-based readiness assessment

How Teams Use This Signal

During operations or training
  • Detect deviation onset while activity is in progress
  • Flag execution drift without halting the exercise
  • Preserve contextual evidence for review
During review and evaluation
  • Compare intended behavior against observed execution
  • Identify recurring deviation patterns across sessions
  • Anchor feedback in observable movement data
Integration snapshot

Integration & Deployment

Designed for direct adoption inside existing mission systems without workflow disruption.

Data Interface
Inputs
  • Planned route definitions (KML, KMZ, GeoJSON)
  • Executed movement telemetry (CSV, JSON, Parquet)
  • Mission metadata and timestamps
Outputs
  • Structured deviation events
  • Deviation magnitude and duration metrics
  • Recovery and non-recovery indicators
  • JSON outputs for downstream systems, with optional PDF mini-after-action extract
Execution & Control
Deployment Models
  • Standalone Python module
  • Containerized microservice
  • API-ready integration
  • Air-gapped compatible
Configuration
  • Doctrine thresholds defined in configuration files
  • Operator-adjustable parameters
  • Core algorithms remain fixed and deterministic across deployments.
Security
Deployment models and licensing are tailored by environment and mission constraints.
View Integration OptionsStart a Pilot Discussion