Aapda Seva Command Center - RoadSoS 🚑

Rank-Optimized Submission for the National Road Safety Hackathon 2026 Organized by: Centre of Excellence for Road Safety (CoERS), RBG Labs, IIT Madras Theme: AI in Road Safety (Applied AI / AI Research) Track Chosen: RoadSoS

Aapda Seva Command Center is an intelligent, real-time emergency response platform designed rigorously against the strict evaluation criteria of the RoadSoS track. By integrating real-time distress signals with location-based resources, it aims to protect lives during the critical "golden hour."

🏆 Hackathon Stage 1 Compliance Checklist

• Primary Requirement: Provides location-based access to trauma centres, ambulances, police, vehicle rescue, and emergency contacts.
• Structured Database Submitted: JSON/NoSQL schema for the predictive models provided in database_models_schema.json.
• 7-Slide Presentation: See PRESENTATION_OUTLINE.md for the exact required slide structure.
• Code Submission (Python Preferred): AI Model relies natively on Python scikit-learn in our decoupled worker node.
• Word Document Details: See WORD_DOC_TEMPLATE.md containing the mandated list of software packages and assumptions.
• Open Models & Free APIs: Utilizes OpenStreetMap (OSM) / Overpass API to fetch contacts globally without proprietary lock-ins.

🚀 Key Aspects Included (As Per Sec 1.3.3)

1. Comprehensive Responders: Instantly fetches the nearest Police Station, hospitals, and ambulance services.
2. Vehicle & On-Road Services: Built-in categorization for tracking Towing services, nearest puncture shops, and showrooms, allowing rapid vehicle rescue during non-medical breakdowns.
3. Global Applicability: The platform does not rely on local, hard-coded data. By utilizing Open APIs (OpenStreetMap), routing and SOS location services function with global applicability across countries.
4. Offline Functionality & Low-Network: The mobile client relies on an offline-first service-worker pattern. If a crash occurs in a cellular dead zone, the payload is cached locally and transmitted immediately upon regaining basic 2G connectivity (ensuring robustness in low-network conditions).

🛠️ Software Stack Highlights

• AI Triage & Database Worker (Python): Python 3.10 with open-source scikit-learn handles the crash severity scoring logic.
• Frontend Core: React 18 (via Vite), TypeScript, Tailwind CSS.
• Database: Firebase Firestore (Real-time synchronization).

💻 Code Execution

1. Install Dependencies: npm install
2. Run the Dashboard: npm run dev
3. Run AI Worker: python data_pipeline/firestore_worker.py (Mock daemon for Python AI processing).

All documentation strictly adheres to the CoERS National Road Safety Hackathon 2026 Rulebook.