AeroLat improves communication reliability in decentralized UAV swarms
AeroLat: Channel-Aware Latent Space Semantic Communication for Decentralized UAV Swarms
Networking and Internet ArchitectureArtificial Intelligence
Summary
When groups of drones need to work together, they share information over wireless links that can be slow or unreliable. The authors noticed that current methods sometimes make all drones send very similar messages, which isn’t helpful. They created AeroLat, a new way for drones to communicate that adjusts messages based on the quality of the connection and keeps the shared information clearer and more distinct. Their tests show AeroLat greatly reduces errors and confusion within the drone swarm.
What this means in practice
- •For drone fleet operators: Enhance coordination among autonomous drone groups operating under limited bandwidth and unreliable wireless links for improved swarm performance.
- •For wireless network engineers: Design communication protocols that adapt to channel conditions to maintain message clarity and reduce false similarity in distributed systems.
Authors
Rajdeep Ghosh, Goparaju Venkata Seshachala Sree Vatsava, Sudip Misra
Abstract
Communication in latent space offers an intriguing alternative to symbolic messages for decentralized autonomous Unmanned Aerial Vehicle (UAV) swarms operating over bandwidth-constrained, time-varying wireless links. However, when homogeneous frozen models are prompted with discretized perceptual inputs, their broadcast states collapse toward the shared prompt template. In view of this, we propose AeroLat, a channel-aware latent semantic communication framework that uses evidence injection. The resulting latent states are then passed through an explicit communication model that encompasses bandwidth-limited serialization, additive noise and information staleness, which facilitates a joint assessment of communication fidelity and swarm-level coordination. Across multi-seed simulations, AeroLat provably remains resilient to codec choice, faults and increasing swarm size. It consistently reproduces the latent-swarm anomaly, while no-whitening controls recover the collapse. In particular, AeroLat is capable of reducing false similarity by 97.5%.