Summary
Communication networks among groups of drones can break up when some drones fail or get lost, making it hard for the whole group to stay connected. The authors created a way for the control station to guess the overall network’s problems using only limited local reports, without needing exact drone locations. They developed a special computer model inspired by physics that helps the drones understand both local and global connections to heal the network. Their system learns from many examples to work faster when new damage happens and keeps drones coordinated while they move to reconnect. Tests with thousands of scenarios showed their method works better than previous approaches.
What this means in practice
- •For drone fleet operators: Manage and restore communication in drone swarms after network failures using a physics-inspired graph model that recovers connections without full location data.
- •For emergency response teams: Coordinate multiple UAV groups in disaster zones by restoring partitioned drone networks quickly to maintain situational awareness and communication.
Authors
Dongbin Jiao, Zhihao Li, Zongsheng Geng, Dongdong Zhao, Shengcai Liu, Shi Yan
Abstract
This paper investigates the resilience and self-healing of unmanned aerial vehicle swarm communication networks under severe network partitioning in complex and adversarial environments. To address limited global observability after damage, we develop a damage-state inference mechanism that enables the ground control station (GCS) to infer global failures from sparse local reports without transmitting location-sensitive coordinates. To recover multi-cluster partitioned topologies, we then propose AdDiff-GT, a physics-inspired graph Transformer that incorporates advection-diffusion dynamics into a node-supernode hybrid graph. Local message propagation captures intra-cluster structural relationships, while global attention models inter-cluster interactions. A weighted composite loss jointly considers connectivity restoration, maneuvering cost, motion consistency, inter-cluster coordination, and spatial preservation during model training. We further develop a meta-learning strategy for transferable initialization across damage patterns and swarm scales, enabling efficient online GCS adaptation and distributed airborne inference. Simulations on 8,000 partitioned scenarios, including 5,000 large-scale cases, show that AdDiff-GT achieves effective topology recovery and spatial preservation over six representative baselines. Meta-learning reduces online planning overhead with a minor increase in recovery steps, balancing prior knowledge transfer and online adaptation.