SIGMA: Sheaf-Informed Geometric Multi-Agent Pathfinding

📄 arXiv: 2502.06440v2 📥 PDF

作者: Shuhao Liao, Weihang Xia, Yuhong Cao, Weiheng Dai, Chengyang He, Wenjun Wu, Guillaume Sartoretti

分类: cs.RO, cs.AI, cs.MA

发布日期: 2025-02-10 (更新: 2025-08-22)

备注: Accepted for presentation at the 2025 IEEE International Conference on Robotics and Automation (ICRA)

期刊: 2025 IEEE International Conference on Robotics and Automation ICRA pp. 1-7

DOI: 10.1109/ICRA55743.2025.11127434

🔗 代码/项目: GITHUB

💡 一句话要点

提出SIGMA框架以解决多智能体路径规划问题

🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture)

关键词: 多智能体路径规划 深度强化学习 层叠理论 去中心化学习 机器人协作

📋 核心要点

  1. 现有的学习型MAPF方法通常依赖于有限的视野，导致智能体在复杂场景中的决策短视和合作效率低下。
  2. 本文提出了一种基于层叠理论的去中心化深度强化学习框架，使智能体通过局部共识学习几何交互依赖关系。
  3. 实验结果显示，所提方法在复杂场景下的路径规划和避碰表现显著优于现有的基线方法。

📝 摘要（中文）

多智能体路径规划（MAPF）问题旨在为多个智能体在已知且可能存在障碍的环境中确定最短且无碰撞的路径。现有的基于学习的方法在处理MAPF问题时，通常依赖于有限的视野，导致短视的决策和低效的合作。为了解决这一挑战，本文提出了一种新的框架，利用层叠理论与去中心化深度强化学习相结合，使智能体能够通过局部共识学习彼此之间的几何交互依赖关系，从而实现紧密的合作决策。实验结果表明，该方法在复杂场景下显著优于现有的学习型MAPF规划器。

🔬 方法详解

问题定义：本文解决的是多智能体路径规划（MAPF）问题，现有方法在复杂环境中由于视野限制，导致智能体决策短视和合作不佳。

核心思路：通过引入层叠理论，本文设计了一种新框架，使智能体能够在局部观察和通信的基础上实现共识，从而提高决策的协同性和效率。

技术框架：整体架构包括智能体的局部观察、共识特征学习和基于深度强化学习的决策模块。智能体通过神经网络在潜在空间中建模共识特征，并进行自监督学习。

关键创新：最重要的创新在于将层叠理论应用于去中心化深度强化学习中，提供了实现全局共识的数学证明，与现有方法相比，显著提升了智能体间的合作能力。

关键设计：在网络结构上，采用了自监督学习机制来训练共识特征，损失函数设计上强调了局部共识的重要性，确保智能体在路径规划中能够有效避免碰撞。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

实验结果表明，所提出的SIGMA方法在多个复杂场景下的路径规划效率提升了显著，尤其在与现有最先进的学习型MAPF规划器对比时，表现出更好的合作与避碰能力，具体性能数据未详述。

🎯 应用场景

该研究的潜在应用领域包括大规模物流、交通运输和机器人部署等场景，能够有效提升多智能体系统在复杂环境中的协作能力和路径规划效率，具有重要的实际价值和未来影响。

📄 摘要（原文）

The Multi-Agent Path Finding (MAPF) problem aims to determine the shortest and collision-free paths for multiple agents in a known, potentially obstacle-ridden environment. It is the core challenge for robotic deployments in large-scale logistics and transportation. Decentralized learning-based approaches have shown great potential for addressing the MAPF problems, offering more reactive and scalable solutions. However, existing learning-based MAPF methods usually rely on agents making decisions based on a limited field of view (FOV), resulting in short-sighted policies and inefficient cooperation in complex scenarios. There, a critical challenge is to achieve consensus on potential movements between agents based on limited observations and communications. To tackle this challenge, we introduce a new framework that applies sheaf theory to decentralized deep reinforcement learning, enabling agents to learn geometric cross-dependencies between each other through local consensus and utilize them for tightly cooperative decision-making. In particular, sheaf theory provides a mathematical proof of conditions for achieving global consensus through local observation. Inspired by this, we incorporate a neural network to approximately model the consensus in latent space based on sheaf theory and train it through self-supervised learning. During the task, in addition to normal features for MAPF as in previous works, each agent distributedly reasons about a learned consensus feature, leading to efficient cooperation on pathfinding and collision avoidance. As a result, our proposed method demonstrates significant improvements over state-of-the-art learning-based MAPF planners, especially in relatively large and complex scenarios, demonstrating its superiority over baselines in various simulations and real-world robot experiments. The code is available at https://github.com/marmotlab/SIGMA