UWB + BLE Hybrid Tracking for Sub-Meter Precision
The challenge
Modern facilities, such as manufacturing floors and hospitals, require varying degrees of asset tracking precision. Certain workflows need sub-meter accuracy (e.g., operating theatres), while the rest of the facility requires low-cost tags and long battery life. Single-technology solutions were either too expensive or too inaccurate, demanding a specialized solution for product engineering.
What we built
EurthTech engineered a hybrid RTLS combining UWB for critical sub-meter positioning and BLE for coarse, long-battery tracking. Tags use a dual-mode AI-powered embedded system where UWB is activated only in high-importance areas, based on BLE region detection. This Edge AI embedded system approach keeps costs down while enabling high-precision tracking where needed, providing a robust backbone for smart lighting systems and general facility management.
How it works
The embedded systems development included tag firmware with adaptive UWB duty cycles based on motion, battery level, and policy. Server-side, our solution features a sophisticated Localization Fusion Engine. This engine is a core GeoAI component, ingesting UWB TDoA fixes, BLE fingerprint cells, and IMU dead-reckoning. We used a factor-graph optimizer to fuse asynchronous fix types and provide a best-estimate trajectory, ensuring AI GIS analytics are highly accurate. UWB anchors were installed with careful geometry and wired sync for the highest precision.
Results
In a 3-floor hospital pilot, the hybrid solution achieved a site-wide median error of around 0.9 m while preserving tag battery life of 6-12 months. Critical zone workflows saw equipment retrieval times drop by 68% and asset utilization increase by 15%, showcasing the tangible benefits of our engineering.
Deployment, Security & Future Scalability: The hybrid approach reduces the need for costly, ubiquitous UWB coverage. Deployment required a careful anchor plan and a policy engine to manage UWB duty. This architecture provides a key layer for a Digital twin smart city strategy, offering the granular data necessary for advanced simulation and Predictive maintenance AI IoT across vast municipal or campus assets.