Skip to content

EDUCATIONAL.md

Latest commit

 

History

History
72 lines (53 loc) · 5.1 KB

EDUCATIONAL.md

File metadata and controls

72 lines (53 loc) · 5.1 KB
Raw

Aethelix: Causal Intelligence for Satellite Fault Management

Aethelix represents a shift from statistical anomaly detection (which asks "is this data weird?") to causal diagnostic reasoning (which asks "why is this happening and what is the physical root cause?").

Why Causal Graphs?

Modern satellites are complex, interconnected systems. A failure in one subsystem (e.g., a power drop) often cascades into others (e.g., thermal fluctuations, software reboots).

Traditional systems use Fixed Thresholds:

  • Simple to implement.
  • Problem: Misses "sub-threshold" faults (e.g., a 5% solar degradation) that are still critical but below the 15% alarm line.
  • Problem: Causes "alarm fatigue" through cascading alerts (one fault triggers 50 alarms).

Aethelix uses Directed Acyclic Graphs (DAGs):

  1. Physics-First: Relationships are derived from spacecraft design, not just data history.
  2. Consolidation: Instead of 50 alarms, Aethelix points to the single root cause that explains all 50 deviations.
  3. Sub-threshold Sensitivity: By summing "weak signals" along causal paths, Aethelix can detect a 5% fault with 90% confidence because the pattern across multiple sensors matches the causal model.

Core Concepts

1. Root Causes

These are the physical failures (e.g., solar_degradation, wheel_friction). They have no parents in the graph.

2. Intermediate States

Unobservable physical states (e.g., battery_efficiency). They help bridge the gap between root causes and sensors.

3. Observables

The telemetry nodes (e.g., battery_voltage_measured). These are mapped to actual sensor data.

4. Bayesian Ranking

Aethelix uses a rule-based Bayesian approach:

  • Posterior Probability: Which cause most likely explains the current set of anomalies?
  • Confidence: How certain are we given the completeness and consistency of the evidence?

Performance Summary

  • Zero-Shot Detection: 100% detection rate on NASA SMAP/MSL dataset without any training.
  • Sub-threshold Advantage: Detects 100% of 5-12% severity faults that traditional 15% thresholds miss entirely.
  • Lead Time: Provides 30-120 seconds of early warning by detecting the "onset" of a fault before it reaches critical limits.

Empirical Flight Validation & DAG Layout Engine (v2.0)

This documentation page includes updates from the v2.0 evaluation suite, incorporating empirical validation on European Space Agency (ESA) flight telemetry and improved hierarchical network rendering:

1. Evaluation on ESA Flight Telemetry (OPS-SAT & ESA-ADB)

  • ESA OPS-SAT (OPSSAT-AD): Evaluated on ADCS reaction wheel magnetic interference and magnetometer attitude anomalies. By applying subsystem-aware persistence filtering ($N=15$ consecutive samples for noisy magnetometer channels, $N=3$ for clean optical sensors), Aethelix achieves a 78.3% F1 score, comparable to deep LSTM autoencoder baselines (78.0% F1) without requiring historical model training.
  • ESA-ADB Multi-Mission: Evaluated on satellite telemetry subject to periodic orbital thermal variation. To prevent false alarms caused by 90-minute day/night baseline oscillations, the engine evaluates innovation residuals ($\Delta x_t = x_t - x_{t-1}$). Decoupling slow orbital dynamics from step-change fault signatures isolates transient subsystem anomalies.

2. Hierarchical Causal DAG Rendering

  • Bounded Functional Corridors: The visualization script (scripts/generate_dag_visuals.py) enforces strict vertical coordinate bounding (maximum node altitude $y=0.68$), ensuring clean separation from structural layer labels ($y=0.87$).
  • Normalized Ranking Score Visualization: In accordance with project semantics, node color intensity maps directly to normalized ranking scores (e.g., 0.94 normalized score weight for PCDU Regulator Failure on GSAT-6A; 0.89 normalized score weight for Reaction Wheel Magnetic Interference on OPS-SAT). Active causal propagation paths connecting dominant root causes to observed telemetry deviations are highlighted along directed edges.

3. CLI Reproduction Commands

Benchmark datasets and diagnostic charts can be generated locally using the standalone scripts:

# Render hierarchical causal DAGs with normalized score weighting
python3 scripts/generate_dag_visuals.py

# Generate empirical validation charts across ESA datasets
python3 scripts/generate_validation_plots.py

# Execute benchmark suite evaluation
python3 scripts/esa_benchmark.py --dataset all

Generated diagnostic charts in docs/:

  • causal_dag_intensity_gsat6a.png — Multi-subsystem causal DAG mapping normalized ranking scores.
  • causal_dag_intensity_opssat.png — ADCS magnetometer interference DAG structure.
  • validation_signal_overlay.png — Telemetry Z-score deviations and persistence window thresholds.
  • validation_confusion_matrix.png — Detection performance comparison against sequence-level baselines.
  • validation_subsystem_metrics.png — Subsystem-level precision, recall, and F1 evaluation.
  • validation_causal_attribution.png — Normalized root-cause score distribution across benchmark scenarios.