Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zhao, Rui, Li, Xihui, Zhang, Yizheng, Liu, Yuzhen, Zhang, Zhong, Zhang, Yufeng, Zhou, Cheng, Zhang, Zhengyou, Han, Lei
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2602.21119
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912924001370112
author Zhao, Rui
Li, Xihui
Zhang, Yizheng
Liu, Yuzhen
Zhang, Zhong
Zhang, Yufeng
Zhou, Cheng
Zhang, Zhengyou
Han, Lei
author_facet Zhao, Rui
Li, Xihui
Zhang, Yizheng
Liu, Yuzhen
Zhang, Zhong
Zhang, Yufeng
Zhou, Cheng
Zhang, Zhengyou
Han, Lei
contents Multi-agent deep Reinforcement Learning (RL) has made significant progress in developing intelligent game-playing agents in recent years. However, the efficient training of collective robots using multi-agent RL and the transfer of learned policies to real-world applications remain open research questions. In this work, we first develop a comprehensive robotic system, including simulation, distributed learning framework, and physical robot components. We then propose and evaluate reinforcement learning techniques designed for efficient training of cooperative and competitive policies on this platform. To address the challenges of multi-agent sim-to-real transfer, we introduce Out of Distribution State Initialization (OODSI) to mitigate the impact of the sim-to-real gap. In the experiments, OODSI improves the Sim2Real performance by 20%. We demonstrate the effectiveness of our approach through experiments with a multi-robot car competitive game and a cooperative task in real-world settings.
format Preprint
id arxiv_https___arxiv_org_abs_2602_21119
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Cooperative-Competitive Team Play of Real-World Craft Robots
Zhao, Rui
Li, Xihui
Zhang, Yizheng
Liu, Yuzhen
Zhang, Zhong
Zhang, Yufeng
Zhou, Cheng
Zhang, Zhengyou
Han, Lei
Robotics
Artificial Intelligence
Multi-agent deep Reinforcement Learning (RL) has made significant progress in developing intelligent game-playing agents in recent years. However, the efficient training of collective robots using multi-agent RL and the transfer of learned policies to real-world applications remain open research questions. In this work, we first develop a comprehensive robotic system, including simulation, distributed learning framework, and physical robot components. We then propose and evaluate reinforcement learning techniques designed for efficient training of cooperative and competitive policies on this platform. To address the challenges of multi-agent sim-to-real transfer, we introduce Out of Distribution State Initialization (OODSI) to mitigate the impact of the sim-to-real gap. In the experiments, OODSI improves the Sim2Real performance by 20%. We demonstrate the effectiveness of our approach through experiments with a multi-robot car competitive game and a cooperative task in real-world settings.
title Cooperative-Competitive Team Play of Real-World Craft Robots
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2602.21119