ORCA 4.0: Universal Autonomous Maritime Operating System

Problem Statement ID: SIH26176 | Theme: Space Technology, Marine Safety & Blue Economy
Sponsoring Agencies: Indian Space Research Organisation (ISRO) & Indian National Centre for Ocean Information Services (INCOIS, Ministry of Earth Sciences)

---

🌊 Executive Summary & Problem Context

India possesses an expansive 7,516 km coastline supporting over 4.5 million artisanal and commercial fishermen. Every day, thousands of small craft venture into deep pelagic waters with critical systemic vulnerabilities:

1. 800+ Annual Fatalities: Traditional marine advisories rely on generalized text broadcasts that ignore vessel-specific hull dimensions, draft, and wave steepness, leading to catastrophic capsizing.
2. ₹45,000 Crore Fuel Wastage: Artisanal boats consume ~40% of their diesel fuel searching blindly for schooling pelagic fish without spatial guidance.
3. Data Silos: ISRO earth observation rasters, INCOIS PFZ bulletins, and Coast Guard maritime radar operate in complete isolation without real-time algorithmic synthesis.
4. Offshore Zero-Connectivity Blackout: Beyond 12 nautical miles, 4G/5G mobile connectivity drops to 0%, rendering standard internet web applications completely non-functional.

ORCA 4.0 is an edge-native, multilingual, physics-grounded Maritime Operating System that bridges space telemetry, continuous mathematical safety evaluations, and low-bandwidth bit-packed offshore radio mesh into a unified, actionable digital twin.

---

🚀 Core Capabilities & Technical Differentiators

1. 🛡️ Continuous Mathematical Risk Engine (< 100ms)

• Zero LLM Hallucination in Safety Decisions: The safety circuit breaker is 100% deterministic and written in pure Python/NumPy mathematics.
• Additive Hazard-Weighted Index (0–100): $$\text{Risk Score} = \sum_{i=1}^n w_i \cdot H_i(\vec{x})$$
  • Wave Height Hazard ($w=0.18$): Real ERA5 swell height relative to vessel length ($H_{\text{crit}} = 0.22 \cdot L + 0.05 \cdot B$).
  • Wave-Vessel Interaction ($w=0.18$): Evaluates wave period, swell direction, and craft heading to detect parametric roll resonance ($\text{Steepness} &gt; 0.35$).
  • Wind & Gale Gusts ($w=0.22$): Real MET Norway 10m wind velocity and squall gust speeds.
  • Surface Currents ($w=0.08$): ROMS ocean velocity vectors.
  • Atmospheric Pressure & Visibility ($w=0.18$): Barometric depressions and fog extinction.
  • Official IMD Warning Override ($w=0.10$): Non-bypassable circuit breaker triggered on active cyclone advisories.

2. 🐟 Multi-Species Bio-Thermal Habitat Suitability Index (HSI)

• Ingests ISRO Oceansat-3 (OCM-3) 360m Chlorophyll-a and INSAT-3DR Sea Surface Temperature (SST) thermal fronts.
• Dynamic Gaussian species probability models for commercial pelagic species:
  • Bangda (Indian Mackerel): Optimal SST 27.5–29.2°C, Chl 1.2–3.0 mg/m³
  • Surmai (Kingfish / Seer Fish): Optimal SST 26.5–28.5°C, Chl 0.8–2.5 mg/m³
  • Tarli (Indian Oil Sardine): Optimal SST 27.0–29.0°C, Chl 2.0–4.5 mg/m³
  • Poplet (Pomfret): Optimal SST 25.0–28.0°C, Chl 1.0–2.8 mg/m³

3. 🧭 Multi-Objective Pareto Routing & Fuel Optimization

• Solves 3 simultaneous Pareto-optimal nautical trajectories:
  1. Safest Detour: Maximum clearance from breaking swells, shallow reefs, and naval restricted zones.
  2. Lowest Fuel: Direct hydrodynamic line minimizing propeller slip and Brake Specific Fuel Consumption (BSFC).
  3. Highest Net Value: Trajectory passing through high-density Chlorophyll front pelagic schooling zones.

4. 💰 Wholesale Harbor Price Arbitrage

• Ingests live wholesale auction rates across 12+ Indian coastal harbors (e.g., Mirkarwada, Sassoon Dock Mumbai, Panaji, Veraval, Mangalore, Kochi).
• Recommends the profit-maximizing landing center by computing: $$\text{Net Profit} = (\text{Catch Weight} \times \text{Auction Price}) - \text{Fuel Cost}(\text{Distance}) - \text{Risk Penalty}$$

5. 📡 16-Byte Bit-Packed LoRa & NavIC Protocol

• Sub-gigahertz telemetry protocol designed for transmission over 868MHz LoRa mesh and ISRO NavIC / Nabhmitra transceivers:
  • Byte 0: Packet Type & SOS Header
  • Bytes 1–3: Latitude (millidegree precision)
  • Bytes 4–6: Longitude (millidegree precision)
  • Byte 7: Deterministic Safety Risk Score ($0-100$)
  • Bytes 8–11: Target PFZ Waypoint Coordinate
  • Bytes 12–15: HMAC-SHA256 Cryptographic Integrity Signature

6. 🛰️ Coast Guard Institutional Governance & SAR Hub

• 1,000-Particle Monte Carlo SAR Drift Engine: Models Stokes wave drift, 3% wind leeway, and ocean currents to generate Search & Rescue probability search ellipses.
• Sentinel-1 Dark Fleet Radar Matcher: Cross-correlates Synthetic Aperture Radar (SAR) backscatter contacts against live AIS feeds to detect unregistered illicit foreign trawlers.
• Immutable Human Override Ledger: SQLite WAL audit logging for Coast Guard zone closures and emergency maritime geofences.

---

🏗️ 16-Agent Directed Acyclic Graph (DAG) Architecture

┌────────────────────────────────────────────────────────────────────────────────────────┐
│                        LAYER 1: SATELLITE & SENSOR INGESTION                           │
│  • ISRO Oceansat-3 (OCM-3)   • INSAT-3DR Geostationary SST   • Sentinel-1 C-Band SAR   │
│  • Open-Meteo Marine (ERA5)  • MET Norway Locationforecast   • INCOIS ERDDAP Ingest    │
│  • 2D Spatial DINEOF Satellite Cloud Gap Reconstruction Algorithm                      │
└───────────────────────────────────────────┬────────────────────────────────────────────┘
                                            │ Typed Async Event Bus
                                            ▼
┌────────────────────────────────────────────────────────────────────────────────────────┐
│                     LAYER 2: SPATIO-TEMPORAL WORLD MODEL ENGINE                        │
│  • Vessel Digital Twin Hydrodynamics (H_crit = 0.22*L + 0.05*B)                        │
│  • Uber H3 Res-7 Hexagonal Spatial Index (~1.2 km² cell resolution)                   │
│  • IMBL 5NM Boundary Buffer & Naval Range Polygon Geofence Engine                      │
└───────────────────────────────────────────┬────────────────────────────────────────────┘
                                            │
                                            ▼
┌────────────────────────────────────────────────────────────────────────────────────────┐
│                     LAYER 3: SPECIALIZED MICROSERVICE PIPELINE                         │
│  • Deterministic Safety Circuit Breaker    • Multi-Species Bio-Thermal HSI Matrix      │
│  • Multi-Objective A* Pareto Navigator     • Real-Time Wholesale Harbor Arbitrage      │
│  • 1,000-Particle Monte Carlo SAR Drift    • Dark Fleet Radar Anomaly Correlator       │
│  • Multilingual Natural Language Voice Synthesizer in 9 Indian Dialects                │
└───────────────────────────────────────────┬────────────────────────────────────────────┘
                                            │
                                            ▼
┌────────────────────────────────────────────────────────────────────────────────────────┐
│                        LAYER 4: EDGE DELIVERY & CLIENT RUNTIME                         │
│  • React 18 + Vite PWA (Offline Vector Tiles & Service Worker Cache)                   │
│  • Interactive 3D Ocean Digital Twin Globe + Leaflet Nautical Vector Map               │
│  • Web Speech API Real-Time Regional Voice AI (Marathi, Hindi, Gujarati, Tamil, etc.) │
│  • 16-Byte Bit-Packed LoRa / NavIC Telemetry Broadcast Node                            │
└────────────────────────────────────────────────────────────────────────────────────────┘

---

🌍 Global Maritime Harbors Dataset

ORCA 4.0 provides built-in coordinates, tidal baselines, and wholesale auction data for ~50 major harbors across all continents:

• India (West Coast): Malvan, Mirkarwada (Ratnagiri), Panaji (Goa), Sassoon Dock (Mumbai), Veraval, Mangalore, Thoppumpady (Kochi), Kanyakumari.
• India (East Coast & Islands): Royapuram (Chennai), Visakhapatnam, Paradip, Phoenix Bay (Port Blair, Andaman).
• International Hubs: Tokyo Bay, Sydney Harbor, Cape Town, Reykjavík, New York Harbor, Rio Grande, Dubai Port Rashid, Salalah, Singapore, Busan, Rotterdam, and more.

---

🧪 Verification Benchmarks & Reliability

• PyTest Automated Test Suite: 67 / 67 tests passing (100%)
  • Continuous risk engine monotonicity and reproducibility tests
  • Deterministic safety circuit breaker capsize & gale override tests
  • Bayesian SAR sighting trajectory update tests
  • HMAC-SHA256 telemetry pack/unpack verification
  • Geofence IMBL border violation detection tests
• Frontend Production Bundle: Compiled with 0 TypeScript errors (100% clean production build).
• Sub-100ms Performance: Pipeline execution measured at 34–48 ms on standard edge hardware.

---

⚡ Quickstart & Zero-Config Deployment

1. System Requirements

• Python: 3.10 or newer
• Node.js: 18+ and npm

2. Quick Launch

# Clone repository
git clone https://github.com/subhamsje/orca.git
cd orca

# Start with single command (macOS / Linux)
chmod +x start-orca.sh
./start-orca.sh

3. Manual Step-by-Step Setup

Backend (FastAPI):

cd backend
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --host 0.0.0.0 --port 8000 --reload

Frontend (React 18 + Vite):

cd frontend
npm install
npm run dev -- --host 0.0.0.0 --port 5173

• Frontend Application: http://localhost:5173
• Backend Swagger API Docs: http://localhost:8000/docs

---

📂 Repository Structure

orca/
├── backend/
│   ├── data_providers/          # Real-time data connectors (MET Norway, Open-Meteo, NDBC)
│   ├── database/                # SQLite WAL repository & audit ledger
│   ├── risk_engine/             # Continuous mathematical risk engine & circuit breakers
│   ├── services/                # 16 Specialized autonomous agents & microservices
│   │   ├── safety_service.py    # Deterministic hull stability & circuit breaker
│   │   ├── pfz_service.py       # Multi-species HSI bio-thermal matrix
│   │   ├── optimization_engine_service.py # Pareto route optimizer
│   │   ├── economic_service.py  # Wholesale harbor auction arbitrage
│   │   ├── sar_drift_service.py # 1,000-particle Monte Carlo SAR drift
│   │   ├── dark_fleet_service.py# Sentinel-1 SAR vs. AIS radar anomaly matcher
│   │   └── nlg_service.py       # Multilingual natural language generation
│   ├── tests/                   # 67 Unit, physics, and integration test suites
│   ├── utils/                   # Hydrodynamic vessel twins, H3 spatial, LoRa bit-packers
│   └── main.py                  # Primary FastAPI REST & WebSocket server
├── frontend/
│   ├── public/                  # PWA Manifest, Service Worker, vector icons
│   ├── src/
│   │   ├── components/          # 3D Ocean Globe, Voice Assistant, Harbor Directory
│   │   ├── map/                 # Leaflet Map workspace, layers, feature drawer
│   │   ├── ui/orca/             # High-performance nautical glassmorphism primitives
│   │   ├── utils/               # Typed API client, global harbor catalog, formatters
│   │   ├── App.tsx              # Fullscreen fluid marine workspace
│   │   ├── ErrorBoundary.tsx    # Crash protection & recovery boundary
│   │   └── main.tsx             # React mount & Service Worker manager
│   └── vite.config.ts           # Vite bundler & PWA configuration
├── ORCA_4.0_SIH_National_Finals_Deck.pptx # Downloadable 16:9 SIH Finals Pitch Deck
├── SIH_Presentation_Master_Deck.md       # Slide-by-slide master presentation spec
├── CHANGELOG.md                          # Complete engineering changelog
└── README.md                             # Project documentation

---

🏆 Smart India Hackathon (SIH 2026) Pitch Assets

• 📥 Physical PowerPoint Deck: ORCA_4.0_SIH_National_Finals_Deck.pptx
• 📄 Master Presentation Spec & Jury Defense: SIH_Presentation_Master_Deck.md

---

📜 License & Acknowledgments

Developed for the Smart India Hackathon (SIH 2026) under Problem Statement SIH26176.
Sponsored by the Indian Space Research Organisation (ISRO, Department of Space) in collaboration with INCOIS (Ministry of Earth Sciences) and the India Meteorological Department (IMD).