Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 18, 2026Electronics LettersOpen Access

Heterogeneity‐Aware Asynchronous Federated Learning for UAV Cooperative Perception

View Full Paper
Ask AI
Bookmark
Share

Authors

JMJimei MaoJFJing Fan

Discussion

Loading...

Member takes

Overview

Randomized trial demonstrates improved accuracy and participation in UAV systems, suggesting better cooperative perception techniques.

Key Points

  • To address the challenges of system heterogeneity and non-IID data in UAV cooperative perception through an asynchronous federated learning framework.
  • Proposed a heterogeneity-aware asynchronous federated learning framework (HAFL) with elastic aggregation windows.
  • Implemented a hierarchical aggregation mechanism to group UAV nodes and applied gradient pruning and amplification.
  • Introduced a dynamic edge-weight penalty considering model staleness, local accuracy, and participation frequency.
  • HAFL improves classification accuracy by up to 3.27%.
  • Reduces system cost by up to 23.2%.
  • Increases average per-client participation rate by 0.21.

Cite This Study

Mao et al. (2026) studied this question.

synapsesocial.com/papers/6a0aace55ba8ef6d83b7055fhttps://doi.org/10.1049/ell2.70608
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1UAV-assisted Unbiased Hierarchical Federated Learning: Performance and Convergence Analysis2024
  2. 2UAV-Enabled Asynchronous Federated Learning2024
  3. 3A Cost-Efficient Aggregation Strategy for Federated Learning in UAV Swarm Networks Under Non-IID Data2025 · 1 citations
  4. 4Participant and Sample Selection for Efficient Online Federated Learning in UAV Swarms2024 · 21 citations
  5. 5Federated Learning with Accurate Model Training and Low Communication Cost in Heterogeneous Scenarios2024