Fleet Telematics Platform

Part of the worlds-biggest-software-project initiative.

An open-source, AI-native platform that turns raw vehicle telemetry into actionable fleet intelligence -- driver safety scoring, predictive maintenance, fuel analytics, and compliance reporting from a single source of truth.

Fleet Telematics Platform collects continuous vehicle and driver data through OBD-II, J1939, and OEM telematics APIs, processes it in a cloud backend, and surfaces operational insights through web and mobile dashboards. It is built for fleet operators, safety managers, and maintenance teams who need real-time tracking, geofencing, driver behaviour analysis, fuel optimisation, and regulatory compliance without the vendor lock-in and opaque pricing of incumbent solutions.

Why Fleet Telematics Platform?

Opaque pricing and long contracts lock operators in. Samsara requires 36-month commitments with no early exit. Teletrac Navman, Verizon Connect, and Fleet Complete all use quote-only pricing with no self-serve option. Smaller fleets -- 97% of which have fewer than 50 vehicles -- are priced out or locked into unsuitable contracts.
No data portability between providers. There is no widely adopted standard for exporting historical telematics data. Switching providers means losing years of operational history, creating artificial lock-in beyond the contract term.
AI safety models are opaque. Incumbent platforms produce driver risk scores from deep learning models without explaining why a score changed. Fleet managers and drivers cannot trust or act on scores they do not understand.
Driver privacy is an afterthought. No major platform natively surfaces GDPR-compliant data minimisation, driver-controlled data access, or works-council-ready audit trails -- critical requirements in European and unionised environments.
Existing open-source alternatives are incomplete. Traccar (Apache 2.0) handles GPS tracking well across 2,000+ devices but lacks AI coaching, ELD/HOS compliance, predictive maintenance, and DTC analytics. OpenRemote (AGPLv3) requires significant customisation for production fleet use and its copyleft licence prevents proprietary deployment.

Key Features

Real-Time Tracking and Geofencing

GPS location tracking with 30-second or better refresh and persistent map view
Polygon and radius geofence creation with entry/exit event alerts
Dwell-time logging per zone for yard management and customer site analytics
Offline position caching on mobile devices with reliable sync on reconnect

Driver Behaviour and Safety

Per-trip and per-driver scoring across speeding, harsh braking, rapid acceleration, and cornering
AI-powered coaching with natural-language score explanations and personalised improvement plans
Route-aware scoring that accounts for road type, traffic, and route difficulty
Anomalous route detection flagging deviations without manually defined geofences

Predictive Maintenance and Vehicle Health

OBD-II and J1939 data ingestion for fuel consumption, engine RPM, DTC fault codes, and odometer
ML-driven predictive maintenance replacing threshold-based alerting with failure prediction before fault codes appear
Preventive maintenance scheduling by mileage, engine hours, or calendar date

Fuel Analytics and Efficiency

Idling time detection and consumption benchmarking across the fleet
Routing optimisation suggestions based on historical fuel data
Support for mixed EV and ICE fleets with unified energy dashboard (backlog)

Compliance and Reporting

ELD/HOS compliance module targeting FMCSA 49 CFR Part 395
DVIR (Driver Vehicle Inspection Report) workflows
Automated compliance narrative generation from raw HOS and DVIR data
Role-based access control with GDPR-compliant data minimisation and driver consent workflows

Integration and Extensibility

REST API (OpenAPI 3.0 specification) with webhook support
Designed for integration with TMS, ERP, fuel card, and dispatch systems
Natural-language query interface for fleet status and exception reporting

AI-Native Advantage

The platform replaces static threshold-based alerting with machine learning models trained on engine sensor time-series to predict component failures before DTC codes appear. Driver coaching adapts content and frequency to individual learning response rather than applying uniform rule thresholds. A natural-language query interface lets operators ask plain-English questions about fleet status, trends, and exceptions without navigating complex dashboard filters. Every AI-generated score includes a plain-language explanation of the factors that influenced it, addressing the trust gap that exists across all incumbent platforms.

Tech Stack & Deployment

The platform targets self-hosted, cloud, and hybrid deployment modes. Vehicle data is ingested through OBD-II and J1939 hardware dongles or native OEM telematics APIs over cellular connections, with store-and-forward buffering for connectivity gaps. The API layer follows the OpenAPI 3.0 specification. The NMFTA Open Telematics API (available at nmfta-repo/nmfta-opentelematics-api) is a candidate standard for TSP-carrier data exchange. Mobile apps target iOS and Android with offline-first architecture.

Market Context

The global fleet telematics market was valued at $93.61 billion in 2025 and is projected to nearly double by 2034, growing at an 8.70% CAGR. By 2026, 79% of mid-to-large commercial fleets use telematics data for operations management. Enterprise incumbents like Samsara charge $40--50+/vehicle/month for premium features (AI dashcam, coaching), with 36-month lock-in contracts. The primary buyers are fleet operations managers, safety directors, and maintenance supervisors at commercial fleet operators ranging from 10-vehicle local fleets to multi-thousand-vehicle enterprises.

Project Status

This project is in the research and specification phase. Contributions, feedback, and domain expertise are welcome.

Contributing

We welcome contributions from developers, domain experts, and potential users. See CONTRIBUTING.md for guidelines.

Important: All contributions must be your own original work or clearly attributed open-source material with a compatible licence. Copyright infringement and licence violations will not be tolerated and will result in immediate removal of the offending contribution. If you are unsure whether a piece of code, text, or other material is safe to contribute, open an issue and ask before submitting.

Licence

Licence to be determined. See discussion for context.