New drone routing system detects spoofed Remote ID to prevent collisions
Drones can now sniff out fake location signals from nearby aircraft in real time.
This paper presents a decentralized trajectory planner that treats Remote ID broadcasts as untrusted. The system uses received signal-strength measurements from neighboring aircraft to detect spoofing and probabilistically localize a spoofing agent. A chance-constrained formulation converts the resulting uncertainty into a risk-bounded unsafe region, integrated into a per-agent Markov decision process-based planner. Simulations of multi-aircraft package delivery show reduced near mid-air collision events compared to planners that assume truthful Remote ID data, while maintaining computational efficiency suitable for real-time execution.
- Treats Remote ID broadcasts as untrusted, using RSS measurements to detect spoofing
- Converts spoofing uncertainty into risk-bounded no-fly zones via chance-constrained MDP
- Simulations show reduced near mid-air collisions in package delivery while keeping real-time performance
Why It Matters
As drone swarms enter commercial airspace, this research provides a practical defense against malicious location spoofing that could cause collisions.