MAGNNET: Multi-Agent Graph Neural Network-based Efficient Task Allocation for Autonomous Vehicles with Deep Reinforcement Learning

TL;DR

This paper presents MAGNNET, a multi-agent graph neural network framework combined with deep reinforcement learning for efficient, decentralized task allocation among autonomous vehicles, achieving high success rates and scalability in complex environments.

Contribution

Introduces a novel GNN-based MARL framework with PPO for decentralized task allocation, improving efficiency and scalability in multi-agent autonomous systems.

Findings

92.5% conflict-free success rate

7.49% performance gap compared to centralized methods

Scalable to 20 agents with 2.8 s processing time

Abstract

This paper addresses the challenge of decentralized task allocation within heterogeneous multi-agent systems operating under communication constraints. We introduce a novel framework that integrates graph neural networks (GNNs) with a centralized training and decentralized execution (CTDE) paradigm, further enhanced by a tailored Proximal Policy Optimization (PPO) algorithm for multi-agent deep reinforcement learning (MARL). Our approach enables unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) to dynamically allocate tasks efficiently without necessitating central coordination in a 3D grid environment. The framework minimizes total travel time while simultaneously avoiding conflicts in task assignments. For the cost calculation and routing, we employ reservation-based A* and R* path planners. Experimental results revealed that our method achieves a high 92.5% conflict-free success rate, with only a 7.49% performance gap compared to the centralized Hungarian method, while outperforming the heuristic decentralized baseline based on greedy approach. Additionally, the framework exhibits scalability with up to 20 agents with allocation processing of 2.8 s and robustness in responding to dynamically generated tasks, underscoring its potential for real-world applications in complex multi-agent scenarios.