Constellation Design Skill

Read CONVENTIONS.md at the repo root before proceeding.

This skill designs and evaluates satellite constellations — multiple spacecraft working together to provide coverage, capacity, or capability that a single satellite cannot. Constellation design is fundamentally different from single-satellite design: the system-level architecture (number of planes, phasing, altitude) drives individual satellite requirements, not the other way around.

Before You Begin

Ask the user (if not already known):

1. Mission objective? (Communications, Earth observation, weather, navigation, IoT, SSA, science — each has different coverage metrics)
2. Coverage requirement? (Continuous global, regional, specific latitude bands, revisit time, max gap duration)
3. Number of satellites (range)? (Budget-driven constraint — can be 3 or 30,000)
4. Inter-satellite links? (Yes = mesh network, reduced ground stations. No = simpler satellites, more ground infrastructure)
5. What design phase?

Applicable Phases

• Primary: Phase A (constellation architecture trade), Phase B (orbit and phasing design)
• Supporting: Phase C (deployment sequencing), Phase D (on-orbit constellation management)

Core Concepts

1. Walker Constellation Notation

The standard notation for symmetric constellations: T/P/F

• T: Total number of satellites
• P: Number of orbital planes (equally spaced in RAAN)
• F: Phasing parameter (relative spacing between planes, 0 ≤ F < P)
• Example: GPS = 24/6/1 at 55° inclination, 20,200 km altitude
• Example: Iridium = 66/6/1 at 86.4° inclination, 780 km altitude

2. Coverage Geometry

• Half-cone angle (footprint): $\rho = \arccos(R_E / (R_E + h))$ — the angular radius of visibility from one satellite.
• Minimum elevation angle ($\epsilon_{min}$): Typical 5-15°. Higher elevation = smaller footprint but better link quality.
• Effective footprint: $\theta = 90° - \epsilon_{min} - \arcsin((R_E \cdot \cos\epsilon_{min}) / (R_E + h))$
• Street-of-coverage width: For a single plane, the ground swath covered by N satellites evenly spaced.

3. Coverage Metrics

Metric	Definition	Typical Target
Continuous coverage	100% of target area has ≥1 satellite visible at all times	GPS, comms constellations
Revisit time	Max time between consecutive passes over a point	EO: 1-24 hours
Max gap	Longest period with no coverage at any point	Comms: 0 (continuous)
Number of folds	Minimum simultaneous visible satellites at any point	Navigation: ≥4 for 3D fix
Contact duration	Time a satellite is visible per pass	LEO: 5-15 min per pass

4. Key Trade Parameters

Parameter	Lower Value	Higher Value
Altitude	Lower drag, shorter life, smaller footprint, more sats needed	Higher footprint, fewer sats, radiation, higher launch ΔV
Inclination	Covers equatorial region well, misses poles	Near-polar covers all latitudes, ground track complexity
N satellites	Cheaper, less coverage, longer revisit	Better coverage, higher cost, more complex management
N planes	Simpler deployment (all in one plane)	Better longitudinal distribution, needs multiple launches or RAAN drift

Analysis Workflows

1. Coverage Analysis

• Analytical (Phase A): Use the Walker formula to estimate single-fold or multi-fold coverage for a given T/P/F/i combination.
• Grid-based: Discretize the Earth's surface and compute visibility from each satellite at time steps.
• Higher fidelity: Recommend STK, GMAT, or Orekit propagation for detailed gap analysis.
• Latitude dependency: Coverage is always better near the inclination-matching latitude. Polar coverage requires i > 80°.

2. Inter-Satellite Links (ISL)

• Intra-plane links: Between adjacent satellites in the same plane. Geometry is relatively stable.
• Cross-plane links: Between satellites in adjacent planes. Range and angle vary — most complex for seam planes (where ascending/descending nodes meet).
• RF ISL: Proven (TDRS heritage), lower data rate (1-10 Gbps typical).
• Optical ISL: Higher data rate (10-100+ Gbps), narrower beam, needs precision pointing (Starlink V2 uses optical ISL).
• Latency benefit: ISL routing can be faster than fiber for long distances — signal travels at c in vacuum vs. ~0.67c in fiber.

3. Deployment Strategy

• Shared launch: Multiple satellites per launch to the same plane. Then phase within the plane using differential drag or low-thrust maneuvering.
• RAAN spreading: Either launch to different RAANs directly (expensive) or use J2 precession at different altitudes to naturally drift planes apart.
• Build-up: Deploy in phases — initial operating capability (IOC) with partial constellation, full operating capability (FOC) with all satellites.

4. Constellation Management

• Station-keeping: Drag makeup (LEO), orbit maintenance, RAAN/argument-of-perigee corrections.
• Spare strategy: On-orbit spares (parked at different altitude), ground spares, or rapid launch capability.
• Collision avoidance: Maneuver authority, conjunction assessment cadence, coordination with other operators.
• End-of-life: Deorbit within 5 years of mission end (current guidelines, moving toward 0 years). $\Delta V_{deorbit}$ budget.

5. Spectrum & Regulatory

• ITU coordination: Frequency filing, interference analysis with existing systems.
• Orbital debris compliance: FCC/ITU 25-year rule (legacy), accelerated timelines for large constellations.
• Licensing: National licensing (FCC for US, Ofcom for UK, etc.) before deployment.

Output Format

1. Constellation Design Report (constellation_report.md): Walker notation, orbit parameters, coverage analysis, ISL architecture.
2. Coverage Map / Statistics: Revisit time, max gap, number of folds by latitude.
3. Deployment Plan: Launch manifest, phasing strategy, IOC/FOC timeline.
4. Per-Satellite Requirements: Derived requirements for each satellite (mass, power, propulsion for station-keeping and deorbit).
5. 🟢 / 🟡 / 🔴 status: Coverage compliance, regulatory readiness, collision risk.

Interface

• Reads from: /requirements/, /analysis/mission-analysis-specialist/ (orbital mechanics, ΔV), /analysis/communications-assessment/ (link budgets for ISL and ground), /analysis/cost-modeling/ (per-satellite and total constellation cost)
• Writes to: /analysis/constellation-design/
• Consumed by: systems-engineering-assessment (per-satellite requirements flowdown), cost-modeling (constellation economics — per-sat × quantity), trade-study-manager (constellation architecture as a trade option)