AeroGuard

A dock-based drone response system for smart homes. Launches autonomously when perimeter sensors detect anomalies.

Stack

Layer	Technology
Frontend	Next.js 14 (App Router) + TypeScript + Tailwind CSS
Backend	Node.js + Express
Database	MongoDB Atlas (Mongoose)
AI Analysis	Google Gemini (multimodal)
Voice Alerts	ElevenLabs TTS
Daily Summary	MeDo (via Gemini)
Real-time	Server-Sent Events (SSE)

MVP Features

1. Perimeter sensor triggers anomaly (motion, door, camera, thermal)
2. Decision engine dispatches drone or marks false alarm
3. Frontend visualises drone launching from dock with live phase updates
4. Gemini analyses camera + thermal feed images
5. Structured threat analysis (level, objects detected, recommended action)
6. Dashboard alert + MongoDB event log + ElevenLabs voice broadcast

Project Structure

AeroGuard/
├── backend/          Node.js + Express API
│   ├── config/       DB, constants
│   ├── controllers/  Route handlers
│   ├── middleware/   Error handling, rate limiting
│   ├── models/       Mongoose schemas
│   ├── routes/       Express routers
│   ├── services/     Gemini, ElevenLabs, SSE, mission simulation
│   └── utils/        Logger, image processing, async wrapper
└── frontend/         Next.js app
    ├── app/          App Router pages
    ├── components/   React components
    ├── hooks/        useSSE, useMission, useVoiceAlert
    └── lib/          Types, API client, AppStateContext

Setup

1. Backend

cd backend
npm install
cp .env.example .env
# Fill in MONGODB_URI, GEMINI_API_KEY, ELEVENLABS_API_KEY, ELEVENLABS_VOICE_ID
npm run dev

2. Frontend

cd frontend
npm install
cp .env.local.example .env.local
# NEXT_PUBLIC_API_URL=http://localhost:5000
npm run dev

Environment Variables

Backend (backend/.env)

Variable	Description
MONGODB_URI	MongoDB Atlas connection string
GEMINI_API_KEY	Google AI Studio API key
GEMINI_MODEL	Model name (default: gemini-1.5-flash)
ELEVENLABS_API_KEY	ElevenLabs API key
ELEVENLABS_VOICE_ID	ElevenLabs voice ID
PORT	Server port (default: 5000)
ALLOWED_ORIGINS	Comma-separated CORS origins

Frontend (frontend/.env.local)

Variable	Description
NEXT_PUBLIC_API_URL	Backend base URL
NEXT_PUBLIC_SSE_URL	SSE stream URL

API Reference

Method	Path	Description
POST	/api/sensors/trigger	Fire a sensor anomaly
GET	/api/missions/active	Get active drone mission
GET	/api/events	List events (paginated)
GET	/api/events/stats/today	Today's activity counts
POST	/api/analysis/run	Run Gemini analysis
POST	/api/voice/broadcast	Synthesise + broadcast voice alert
GET	/api/summary/today	Get/generate today's MeDo summary
GET	/api/stream	SSE real-time event stream

Data Flow

Sensor Trigger
    ↓ POST /api/sensors/trigger
Decision Engine (missionDecisionService)
    ↓ shouldDispatch = true
Create Mission → Start Simulation
    ↓ SSE: mission:created, mission:phase (every few seconds)
Frontend DroneDocWidget animates phase transitions
    ↓
POST /api/analysis/run (with camera + thermal base64)
    ↓ Gemini multimodal API
Structured JSON analysis stored on Event
    ↓ SSE: gemini:complete
AlertBanner shows threat + POST /api/voice/broadcast
    ↓ ElevenLabs TTS → audio file
VoiceAlertPlayer auto-plays audio
    ↓
Nightly cron: MeDo generates daily summary