English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deploying teams of unmanned aerial vehicles (UAVs) to harvest data fromdistributed Internet of Things (IoT) devices requires efficient trajectoryplanning and coordination algorithms. Multi-agent reinforcement learning (MARL)has emerged as a solution, but requires extensive and costly real-worldtraining data. To tackle this challenge, we propose a novel model-aidedfederated MARL algorithm to coordinate multiple UAVs on a data harvestingmission with only limited knowledge about the environment. The proposedalgorithm alternates between building an environment simulation model fromreal-world measurements, specifically learning the radio channelcharacteristics and estimating unknown IoT device positions, and federated QMIXtraining in the simulated environment. Each UAV agent trains a local QMIX modelin its simulated environment and continuously consolidates it through federatedlearning with other agents, accelerating the learning process. A performancecomparison with standard MARL algorithms demonstrates that our proposedmodel-aided FedQMIX algorithm reduces the need for real-world trainingexperiences by around three magnitudes while attaining similar data collectionperformance.

Model-aided Federated Reinforcement Learning for Multi-UAV Trajectory Planning in IoT Networks