Dynamic Mask Enhanced Intelligent Multi-UAV Deployment for Urban Vehicular Networks

arXiv cs.AI / 4/6/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper targets challenges in urban Vehicular Ad Hoc Networks (VANETs), where frequent link disconnections and subnet fragmentation degrade reliable vehicle-to-network connectivity.
It proposes a dynamic multi-UAV relay deployment strategy to improve connectivity for urban vehicular communications while controlling multi-UAV energy usage.
The core contribution is a Score-based Dynamic Action Mask enhanced QMIX algorithm (Q-SDAM) that uses a score-driven dynamic action masking mechanism to handle large action spaces and speed up learning.
Experiments using real-world datasets indicate Q-SDAM increases vehicle connectivity by 18.2% and cuts multi-UAV energy consumption by 66.6% versus prior algorithms.
The study emphasizes practical viability by validating the approach with realistic data rather than purely simulated settings.

Abstract

Vehicular Ad Hoc Networks (VANETs) play a crucial role in realizing vehicle-road collaboration and intelligent transportation. However, urban VANETs often face challenges such as frequent link disconnections and subnet fragmentation, which hinder reliable connectivity. To address these issues, we dynamically deploy multiple Unmanned Aerial Vehicles (UAVs) as communication relays to enhance VANET. A novel Score based Dynamic Action Mask enhanced QMIX algorithm (Q-SDAM) is proposed for multi-UAV deployment, which maximizes vehicle connectivity while minimizing multi-UAV energy consumption. Specifically, we design a score-based dynamic action mask mechanism to guide UAV agents in exploring large action spaces, accelerate the learning process and enhance optimization performance. The practicality of Q-SDAM is validated using real-world datasets. We show that Q-SDAM improves connectivity by 18.2% while reducing energy consumption by 66.6% compared with existing algorithms.