WIP AGPL-3.0Insurance coverage

Telematics Hub

Kafka OBD-II Telematics Pipeline Open-Source for Insurance and Fleet Management

Status
WIP
License
AGPL-3.0
Category
Insurance coverage
Updated
GitHub source

Project status

Description

Telematics Hub is an open-source platform for ingesting and analyzing telematics data from connected vehicles (OBD-II dongle, integrated telematics OEM, smartphone insurance app), designed for pay-per-use (PAYD/PHYD) insurance use cases and fleet management. The architecture uses event-driven Kafka Streams: devices publish MQTT events (GPS position, speed, acceleration, braking, RPM motor, fuel consumption, OBD-II error codes) to an MQTT gateway broker, which forwards them to structured Kafka topics. Kafka Streams calculates real-time aggregate metrics (driving score, harsh braking events, speeding events, daily distance, eco-driving score) and persists them in TimescaleDB for efficient time-series queries. The data model supports cross-OEM normalization (Tesla, Volkswagen Group, Stellantis, BMW, generic MQTT) via a pluggable adapter. Security is critical for sensitive data (vehicle position = driver position): TLS 1.3 encryption to the MQTT broker, device certificate-based authentication, optional GPS anonymization (configurable geohash precision, snap-to-route for privacy), GDPR consent management (explicit driver consent for sharing data with insurance). The analytics module offers: risk scoring based on driving behavior aggregates, fleet manager reporting (vehicle usage, driving hours, operational costs), real-time alerts (probable collision, eco-driving outside limits, predictive maintenance based on OBD-II codes). Licensed under AGPL-3.0 to protect the platform: any fork offering insurance/fleet services to third parties must open-source modifications. Deploy target: dual-purpose insurance (auto + fleet), MGA/intermediary insurers, fleet manager logistics, software house automotive. The philosophy is 'driver data sovereignty': data is not shared with OEM/dealers without explicit consent, and storage is on-premises for the client.

Key Features

  • MQTT Gateway + Kafka Pipeline

    Broker for MQTT (Mosquitto/EMQX) for device ingestion, forward structured Kafka topics, throughput target 10k events/s per node (from benchmark).

  • Real-time Analytics with Kafka Streams

    Harsh Braking Events, Speeding Events, Eco-Drive Score, Sliding Window: 5 min / 1 h / 24 h, State Store (RocksDB)

  • Time Series Database

    Table for device data, compressed in chunks, configurable retention by event type, continuous aggregates for dashboards.

  • OBD-II + GPS for multiple OEMs

    Plug adapter compatible with Tesla, VW Group, Stellantis, BMW, OBD-II generic, and smartphone insurance app SDK for Android and iOS.

  • Encryption (TLS 1.3) + Device Authentication

    TLS 1.3 against MQTT broker, device certificate-based authentication, certificate pinning, automated rotation via internal PKI.

  • GDPR Consent Management

    Explicit consent leading to data sharing, geohash precision configurable (4-12), optional snap-to-route, right to oblivion for the device.

  • Risk Scoring and Reporting

    Algorithmic risk scoring for car insurance policy, fleet manager dashboard (usage, driving hours, costs), real-time collision and maintenance alerts.

  • Optional OEM API integration

    Connector for Tesla API, VW We Connect, Stellantis Connect (requires OEM partner credentials), falls back to generic OBD-II dongle.

Tech stack

  • Kafka Streams
  • Spring Boot 3.4
  • Java 21
  • TimescaleDB
  • MQTT (Mosquitto/EMQX)
  • OBD-II
  • Apache Kafka
  • Docker Compose
  • RocksDB

Use cases

  • Auto insurance that offers pay-per-use (PAYD) or pay-how-you-drive (PHYD) policies with data-driven risk scoring.
  • Fleet Manager Logistics with 50-500 vehicles to monitor usage, operational costs and eco-driving.
  • MGA broker offering innovative insurance solutions to partners.
  • Software house for automotive that integrates telematics into its own apps without relying on SaaS providers US

Road Map

  1. v0.1.0Scheduled

    Core: MQTT, Kafka, and TimescaleDB

    • MQTT Gateway with Certificate-Based Authentication
    • Kafka producer + structured topic
    • TimeScaleDB Hypertable and Ingestion Base
    • OBD II Generic Script for Testing
  2. v0.5.0Scheduled

    Analytics: Kafka Streams and Risk Scoring

    • Kafka Streams performance score + error rate
    • Risk scoring algorithm MVP
    • Continuous aggregates on TimescaleDB
    • Dashboard Fleet and Insurance Overview
  3. v1.0.0Scheduled

    Multi-OEM compliant with GDPR for production deployment

    • Tesla Adapter with VW Group and Stellantis
    • GDPR Consent and Geohash Anonymization
    • Predictive Maintenance from OBD-II Codes
    • Documentation for MkDocs and Compliance Guide

Who is it for

  • For Developer Senior
  • Decision Maker/Founder

Quick start

Start the project locally in just a few commands. 

# Quick start (work-in-progress, repo non ancora pubblicato)
# git clone https://github.com/<tba>/telematics-hub.git
# cd telematics-hub
# docker compose up -d  # mosquitto + kafka + timescale + telematics-hub
# # Pubblica evento test con mosquitto_pub o usa simulator script

Contribute to the project

Telematics Hub is open source. Open an issue, propose a pull request or leave a star on GitHub: every contribution, no matter how small, is welcome.

  • ⭐ Leave a star if it's useful
  • 🐛 Report bugs or suggest features via issue
  • 🔧 Send a pull request with fixes or improvements
  • 📣 Share the project if you find it interesting
Star on GitHubOpen an issuePull request
← ⭐ View on GitHub