Causal-Inspired Multi-Agent Decision-Making via Graph Reinforcement Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Jing, Jin, Yan, Ding, Fei, Wei, Chongfeng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911084788580352
author Wang, Jing
Jin, Yan
Ding, Fei
Wei, Chongfeng
author_facet Wang, Jing
Jin, Yan
Ding, Fei
Wei, Chongfeng
contents Since the advent of autonomous driving technology, it has experienced remarkable progress over the last decade. However, most existing research still struggles to address the challenges posed by environments where multiple vehicles have to interact seamlessly. This study aims to integrate causal learning with reinforcement learning-based methods by leveraging causal disentanglement representation learning (CDRL) to identify and extract causal features that influence optimal decision-making in autonomous vehicles. These features are then incorporated into graph neural network-based reinforcement learning algorithms to enhance decision-making in complex traffic scenarios. By using causal features as inputs, the proposed approach enables the optimization of vehicle behavior at an unsignalized intersection. Experimental results demonstrate that our proposed method achieves the highest average reward during training and our approach significantly outperforms other learning-based methods in several key metrics such as collision rate and average cumulative reward during testing. This study provides a promising direction for advancing multi-agent autonomous driving systems and make autonomous vehicles' navigation safer and more efficient in complex traffic environments.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23080
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Causal-Inspired Multi-Agent Decision-Making via Graph Reinforcement Learning
Wang, Jing
Jin, Yan
Ding, Fei
Wei, Chongfeng
Multiagent Systems
Since the advent of autonomous driving technology, it has experienced remarkable progress over the last decade. However, most existing research still struggles to address the challenges posed by environments where multiple vehicles have to interact seamlessly. This study aims to integrate causal learning with reinforcement learning-based methods by leveraging causal disentanglement representation learning (CDRL) to identify and extract causal features that influence optimal decision-making in autonomous vehicles. These features are then incorporated into graph neural network-based reinforcement learning algorithms to enhance decision-making in complex traffic scenarios. By using causal features as inputs, the proposed approach enables the optimization of vehicle behavior at an unsignalized intersection. Experimental results demonstrate that our proposed method achieves the highest average reward during training and our approach significantly outperforms other learning-based methods in several key metrics such as collision rate and average cumulative reward during testing. This study provides a promising direction for advancing multi-agent autonomous driving systems and make autonomous vehicles' navigation safer and more efficient in complex traffic environments.
title Causal-Inspired Multi-Agent Decision-Making via Graph Reinforcement Learning
topic Multiagent Systems
url https://arxiv.org/abs/2507.23080