A Topology-Aware Spatiotemporal Handover Framework for Continuous Multi-UAV Tracking

Researchers proposed a topology-aware handover system for real-time multi-UAV traffic tracking.

The framework uses geometric overlap and virtual lane topology instead of appearance-based re-identification to keep vehicle IDs consistent across camera views.

In urban tests, it achieved a 99.8% handover success rate, supporting reliable network-level traffic analysis and edge deployment.