Ocearo Core

The First AI Co-pilot for Signal K

Ocearo Core is the voice and brain of the Ocearo ecosystem — an advanced Signal K plugin transforming your vessel into a smart ship. As a true AI Co-pilot, it provides global vessel monitoring, failure prediction, sail trim optimization based on your boat's polar data, and intelligent route planning. All powered by a local LLM (Ollama) and Text-to-Speech output, ensuring privacy and offline capability.

"Just A Rather Very Intelligent System" — Marine Edition 🚢

Overview

Ocearo Core goes beyond simple dashboards. It's an intelligent AI Co-pilot that:

👁️ Monitors — Global surveillance of all vessel data, weather, and AIS in real time
🔮 Predicts — Proactive failure prediction and maintenance alerts before things break
⛵ Optimizes — Sail trim and course optimization matched against your vessel's polar performance
🗺️ Plans — Intelligent route planning and navigation assistance
🗣️ Speaks — Contextual voice feedback and alerts via Piper TTS or eSpeak
🧠 Thinks — Deep contextual analysis with a local LLM (Ollama)
⚓ Anchors — Full anchor management with drag alarms (Signal K Anchor API)
📝 Logs — Automatic logbook with local fallback store + fuel log

Ocearo Ecosystem:

👀 Ocearo-UI — The eyes (3D visual interface)
🗣️ Ocearo-Core — The voice (AI assistant, this plugin)
🧠 Signal K — The nervous system (data bus)

Features

⚓ Anchor Management (Signal K Anchor API)

Drop, raise, reposition anchor via REST endpoints
Configurable alarm radius with drag detection (haversine)
Signal K notifications: notifications.navigation.anchor.drag (emergency) and notifications.navigation.anchor.watch (warn)
Persisted anchor state — survives plugin restarts
Mode-change safety: warns if mode changes while anchor is deployed

� Logbook — Dual Backend

Primary: proxies to @meri-imperiumi/signalk-logbook if installed
Fallback: registers as a Signal K Resource Provider (logbooks) with local JSON store in <dataDir>/ocearo-logbook/
Fuel log always stored locally (fuel-log.json) regardless of backend
AI-enhanced entries via LLM when Ollama is available

🌅 Startup Briefing

Weather forecast, tide times, tank and battery levels
Spoken summary on plugin start

📍 Navigation Points (every 30 min)

Position, speed, course, depth, weather update

⛵ Sail Coaching

Real-time sail trim recommendations
Course optimisation with VMG analysis
Reefing suggestions based on conditions

🚨 Smart Alerts

Intercepts all Signal K notifications
Contextual LLM explanations
Critical alerts announced via TTS
Engine alarm monitoring (notifications.propulsion.*)

🎭 Personalities & Modes

Personas: Captain, Teammate, Jarvis, French Sailor
Modes: sailing, anchored, motoring, moored, racing
Languages: English, French (extensible)

Architecture

plugin/
├── index.js                  # Entry point, Express router, security middleware
├── schema.json               # Admin UI config schema
└── src/
    ├── anchor/
    │   ├── anchor-state.js   # State machine (raised/dropping/dropped/raising)
    │   ├── anchor-alarm.js   # Drag detection + SK notifications
    │   └── anchor-plugin.js  # REST endpoints + registerWithRouter
    ├── analyses/
    │   ├── alert.js          # Alert analysis
    │   ├── ais.js            # AIS collision detection
    │   ├── meteo.js          # Weather analysis
    │   ├── sailcourse.js     # Course optimisation
    │   └── sailsettings.js   # Sail trim recommendations
    ├── brain/
    │   └── index.js          # OrchestratorBrain — schedules, mode, status
    ├── config/
    │   └── index.js          # ConfigManager + i18n
    ├── dataprovider/
    │   ├── signalk.js        # SignalKDataProvider
    │   ├── marineweather.js  # Weather provider
    │   └── tides.js          # Tides provider
    ├── llm/
    │   └── index.js          # LLMClient (Ollama)
    ├── logbook/
    │   ├── index.js          # LogbookManager (dual backend)
    │   └── logbook-store.js  # Local JSON store + Resource Provider
    ├── memory/
    │   └── index.js          # MemoryManager
    └── voice/
        └── index.js          # VoiceModule (Piper / eSpeak / console)

Data Flow

Signal K data bus
      │
      ▼
SignalKDataProvider ──► OrchestratorBrain ──► LLMClient (Ollama)
      │                       │                     │
      │                  Analyzers              VoiceModule
      │                       │                (Piper TTS)
      │                  LogbookManager
      │                 (SK logbook / local store)
      │
      ▼
AnchorPlugin ──► AnchorAlarm ──► SK notifications
             └─► AnchorState (persisted)

Installation

Prerequisites

Signal K Server ≥ 1.x
Node.js ≥ 18.0.0
Ollama (optional, for LLM) — Install Ollama
Piper TTS (optional, for voice) — Install Piper

Install via npm

npm install ocearo-core

Restart Signal K and configure via Admin UI → Server → Plugin Config → Océaro Core.

Install from Source

cd ~/.signalk/node_modules
git clone https://github.com/laborima/ocearo-core.git
cd ocearo-core/plugin
npm install

Configuration

Basic

Setting	Description	Default
`language`	Interface language (`en`/`fr`)	`en`
`persona`	AI personality	`jarvis`
`mode`	Operating mode	`sailing`

Anchor

Setting	Description	Default
`anchor.defaultRadius`	Alarm radius in metres	`30`
`anchor.watchRadiusPercent`	Watch threshold (% of radius)	`80`
`anchor.positionUpdateInterval`	Position check interval (ms)	`2000`

LLM (Ollama)

Setting	Description	Default
`ollamaHost`	Ollama server URL	`http://localhost:11434`
`model`	Model name	`llama3.2:3b`
`timeoutSeconds`	Request timeout	`30`

Voice

Setting	Description	Default
`voice.enabled`	Enable TTS	`true`
`voice.backend`	Engine (`piper`/`espeak`/`console`)	`piper`
`voice.piperModel`	Piper voice model	`en_US-joe-medium`

Scheduling

Setting	Description	Default
`schedules.alertCheck`	Alert check (seconds)	`30`
`schedules.weatherUpdate`	Weather update (seconds)	`300`
`schedules.navPointMinutes`	Navigation point (minutes)	`30`

API Endpoints

All endpoints are under /plugins/ocearo-core/. Rate limits apply (120 req/min general, 10/min for AI operations).

System

Endpoint	Method	Description
`/health`	GET	Component health check
`/status`	GET	Full system status (mode, weather, anchor, logbook backend)
`/analyze`	POST	Trigger AI analysis (`weather`, `sail`, `alerts`, `ais`, `status`, `logbook`, `route`)
`/speak`	POST	Speak text via TTS (`{ text, priority }`)
`/mode`	POST	Change operating mode (`{ mode }`)

Memory

Endpoint	Method	Description
`/memory`	GET	Memory context and statistics
`/memory/stats`	GET	Statistics only
`/memory/context`	POST	Update vessel info / destination

Logbook

Endpoint	Method	Description
`/logbook/all-entries`	GET	All entries (proxied or local)
`/logbook/entries`	GET	Analysis entries only
`/logbook/add-entry`	POST	Add manual logbook entry
`/logbook/entry`	POST	Generate AI-enhanced entry from vessel data
`/logbook/entry`	GET	Retrieve recent AI entries (`?limit=50`)
`/logbook/analyze`	POST	Full AI logbook analysis
`/logbook/stats`	GET	Analysis statistics
`/logbook/fuel`	GET	Fuel log entries
`/logbook/fuel`	POST	Add fuel refill record
`/logbook/backend`	GET	Active backend (`signalk-logbook` or `local`)

Anchor (Signal K Anchor API)

Endpoint	Method	Description
`/navigation/anchor/drop`	POST	Drop anchor at current position
`/navigation/anchor/raise`	POST	Raise anchor
`/navigation/anchor/radius`	POST	Set alarm radius `{ value: metres }`
`/navigation/anchor/reposition`	POST	Reposition `{ rodeLength, anchorDepth }`
`/navigation/anchor/status`	GET	Lightweight status
`/navigation/anchor`	GET	Full anchor state snapshot

LLM

Endpoint	Method	Description
`/llm/test`	POST	Test LLM with custom prompt

Signal K Paths

Subscriptions (Input)

navigation.position, navigation.speedOverGround, navigation.courseOverGroundTrue
navigation.headingTrue, environment.depth.belowKeel
environment.wind.speedApparent, environment.wind.angleApparent
notifications.*

Publications (Output)

notifications.navigation.anchor.drag — drag alarm (emergency)
notifications.navigation.anchor.watch — approaching limit (warn)
notifications.navigation.anchor.modeChange — mode changed while anchored
navigation.anchor.position — anchor drop position
navigation.anchor.currentRadius — active alarm radius
navigation.anchor.maxRadius — configured max radius
navigation.anchor.rodeLength — rode length

Setting Up Dependencies

Ollama (LLM)

# Install Ollama
curl -fsSL https://ollama.ai/install.sh | sh

# Pull a model
ollama pull llama3.2:3b

# Start server
ollama serve

Piper TTS

# Download binary from https://github.com/rhasspy/piper/releases

# English voice
wget https://huggingface.co/rhasspy/piper-voices/resolve/main/en/en_US/joe/medium/en_US-joe-medium.onnx

# French voice
wget https://huggingface.co/rhasspy/piper-voices/resolve/main/fr/fr_FR/tom/medium/fr_FR-tom-medium.onnx

eSpeak (fallback TTS)

# Debian/Ubuntu
sudo apt-get install espeak

# macOS
brew install espeak

Integration with Ocearo-UI

Ocearo Core is designed to work seamlessly with Ocearo-UI:

Anchor controls call /navigation/anchor/* endpoints
Fuel log uses /logbook/fuel with fallback to /logbook/add-entry
AI analysis triggered via /analyze with types weather, sail, alerts, ais, status, logbook, route
Engine alarms read from notifications.propulsion.* Signal K paths
Mode changes propagated via /mode endpoint

Security

Rate limiting — built-in per-IP limiter (no external dependency):
- General: 120 req/min
- AI operations (/analyze, /logbook/entry, /llm/test): 10/min
- TTS (/speak): 20/min
Input sanitisation — control characters stripped, lengths enforced
JSON validation — all POST bodies validated before processing
404 catch-all — unknown routes return structured JSON errors

Contributing

🐛 Report bugs — Open an issue
💡 Suggest features — Share ideas
🔧 Submit PRs — Fix bugs, add features, improve docs
🌍 Translate — Add language support

See CONTRIBUTING.md for guidelines.

Roadmap

Additional languages (Spanish, German, Italian)
More weather providers (NOAA, Météo-France)
Advanced polar performance analysis
Voice command input (speech-to-text)
Autopilot integration
Machine learning for personalised sailing advice

License

Apache License 2.0 — see LICENSE.

Acknowledgments

Signal K — Open marine data standard
Ollama — Local LLM runtime
Piper — Fast local TTS
Ocearo-UI — 3D marine interface
OpenPlotter — Open source sailing platform

Navigation Disclaimer

⚠ Use with Caution – Not a Substitute for Official Navigation Systems

Ocearo Core is designed to enhance sailing awareness and provide intelligent assistance. However, this software is not a certified navigation or safety system and should not be relied upon as the sole source of navigational information.

Always cross-check data with official marine charts, GPS devices, and other navigation aids.
Maintain situational awareness and follow maritime safety regulations.
The developers of Ocearo Core are not liable for any incidents, accidents, or navigation errors that may arise from using this software.

By using Ocearo Core, you acknowledge and accept the inherent risks of relying on non-certified navigation tools. Always navigate responsibly!

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
.github/ISSUE_TEMPLATE		.github/ISSUE_TEMPLATE
docs		docs
plugin		plugin
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.fr.md		README.fr.md
README.md		README.md
build-plugin.sh		build-plugin.sh
deploy.sh		deploy.sh
test.sh		test.sh

Folders and files

Latest commit

History

Repository files navigation

Ocearo Core

Overview

Features

⚓ Anchor Management (Signal K Anchor API)

� Logbook — Dual Backend

🌅 Startup Briefing

📍 Navigation Points (every 30 min)

⛵ Sail Coaching

🚨 Smart Alerts

🎭 Personalities & Modes

Architecture

Data Flow

Installation

Prerequisites

Install via npm

Install from Source

Configuration

Basic

Anchor

LLM (Ollama)

Voice

Scheduling

API Endpoints

System

Memory

Logbook

Anchor (Signal K Anchor API)

LLM

Signal K Paths

Subscriptions (Input)

Publications (Output)

Setting Up Dependencies

Ollama (LLM)

Piper TTS

eSpeak (fallback TTS)

Integration with Ocearo-UI

Security

Contributing

Roadmap

License

Acknowledgments

Navigation Disclaimer

Support

About

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages