SwarmUpdate: Hierarchical Software Updates and Deep Learning Model Patching for Heterogeneous UAV Swarms

TL;DR

SwarmUpdate tackles the challenge of efficiently updating software and DL perception models in heterogeneous UAV swarms. It introduces SwarmSync, a hierarchical, type-based update distribution mechanism, and SwarmModelPatch, a patching approach that freezes selected model layers to minimize update size while preserving accuracy. Through ARGoS simulations, the work demonstrates that SwarmSync outperforms gossip- and auction-based baselines in synchronization efficiency, with manageable overhead, and that SwarmModelPatch can dramatically reduce patch sizes and transmission costs while maintaining acceptable overall and new-class accuracies. The results suggest significant practical benefits for in-field updates in mission-critical, large-scale UAV deployments, and point to future work on security and broader applicability of model patching. Overall, SwarmUpdate provides a foundational framework for robust, scalable, and bandwidth-efficient updates in heterogeneous drone swarms, enabling rapid adaptation to evolving mission data and objectives.

Abstract

Heterogeneous unmanned aerial vehicle (UAV) swarms consist of dozens to hundreds of drones with different roles and varying hardware and software requirements collaborating towards a shared mission. While traditional approaches for synchronized software updates assume swarms to be unstructured and homogeneous, the heterogeneous nature of modern swarms and the emerging need of drones to update their deep learning (perception) models with new objectives or data as a mission unfolds, has made efficient software update methods crucial for swarms to adapt to dynamic environments. To address these challenges, we introduce the SwarmUpdate framework for software updates in heterogeneous UAV swarms, composed of two key components: SwarmSync and SwarmModelPatch. SwarmSync is a hierarchical software update synchronization strategy to distribute a software update to the right subset of drones within a swarm, while SwarmModelPatch is a deep learning model patching method that reduces the size of a (deep learning model) update by only allowing some layers of the model to be updated (freezing the other layers). In this paper, we systematically evaluate the performance of SwarmSync through large-scale simulations in the ARGoS swarm simulator, comparing SwarmSync to auction-based (SOUL) and gossip-based rebroadcasting (Gossip) baselines, and SwarmModelPatch to a non-incremental model patching strategy.

Figures (7)

• Figure 1: Structure of a typical CNN model.
• Figure 2: Swarm UAV formations and roles for SwarmSync and the two baseline software update synchronization strategies.
• Figure 3: 18 Drones surrounding the Updater.
• Figure 4: The average time taken per drone for each synchronization strategy to converge the swarm.
• Figure 5: The average overhead per drone for each synchronization strategy until convergence of the swarm.