MAAIApr 11, 2025

Graph Based Deep Reinforcement Learning Aided by Transformers for Multi-Agent Cooperation

arXiv:2504.08195v16 citationsh-index: 52025 IEEE International Conference on Communications Workshops (ICC Workshops)
Originality Highly original
AI Analysis

This addresses multi-agent coordination for drones in applications like disaster response, with incremental improvements in performance metrics.

The paper tackles mission planning for cooperative autonomous drones in uncertain environments by proposing a framework integrating Graph Neural Networks, Deep Reinforcement Learning, and transformers, achieving 90% service provisioning, 100% grid coverage, and reducing average steps per episode from 600 to 200 compared to benchmarks.

Mission planning for a fleet of cooperative autonomous drones in applications that involve serving distributed target points, such as disaster response, environmental monitoring, and surveillance, is challenging, especially under partial observability, limited communication range, and uncertain environments. Traditional path-planning algorithms struggle in these scenarios, particularly when prior information is not available. To address these challenges, we propose a novel framework that integrates Graph Neural Networks (GNNs), Deep Reinforcement Learning (DRL), and transformer-based mechanisms for enhanced multi-agent coordination and collective task execution. Our approach leverages GNNs to model agent-agent and agent-goal interactions through adaptive graph construction, enabling efficient information aggregation and decision-making under constrained communication. A transformer-based message-passing mechanism, augmented with edge-feature-enhanced attention, captures complex interaction patterns, while a Double Deep Q-Network (Double DQN) with prioritized experience replay optimizes agent policies in partially observable environments. This integration is carefully designed to address specific requirements of multi-agent navigation, such as scalability, adaptability, and efficient task execution. Experimental results demonstrate superior performance, with 90% service provisioning and 100% grid coverage (node discovery), while reducing the average steps per episode to 200, compared to 600 for benchmark methods such as particle swarm optimization (PSO), greedy algorithms and DQN.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes