跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2610.18186 · v2

Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields

面向大规模终身多智能体路径规划的共享空间场强化学习

首次提交 2026-10-13 19:25
版本更新 2026-11-19 01:45
雷达观测 2026-11-19 15:04
arXiv ↗PDF ↗

摘要

Large-scale applications of lifelong multi-agent path finding (LMAPF) call for computationally inexpensive movement decisions under changing traffic conditions and across varied layouts. We propose a navigation policy in which a convolutional network computes a shared spatial field over the map once per timestep and a compact goal-conditioned operator derives agents' movement preferences. The complete policy is trained end-to-end using average-reward reinforcement learning with throughput as the reward, without planner demonstrations or a separate online path planner. For structured and unstructured layouts, we train one policy per class on pools of small maps and deploy each without retraining on larger, unseen layouts and agent populations. Across six layouts with 10,000 agents, our policies achieve throughput comparable to or higher than the strongest evaluated baselines. The proposed method has an average execution time below 10 ms per step, more than an order of magnitude faster than several established planners.

AI 论文解读

基于摘要 · 非全文审阅

提出一种基于共享空间场和平均奖励强化学习的导航策略,无需规划器演示或在线路径规划,即可在大规模终身多智能体路径规划中实现高吞吐量和低计算开销。

正在读取账号与研究记录…

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-19 01:45 · 最新观测版本

Reinforcement Learning for Large-Scale Lifelong Multi-Agent Path Finding with Shared Spatial Fields

Large-scale applications of lifelong multi-agent path finding (LMAPF) call for computationally inexpensive movement decisions under changing traffic conditions and across varied layouts. We propose a navigation policy in which a convolutional network computes a shared spatial field over the map once per timestep and a compact goal-conditioned operator derives agents' movement preferences. The complete policy is trained end-to-end using average-reward reinforcement learning with throughput as the reward, without planner demonstrations or a separate online path planner. For structured and unstructured layouts, we train one policy per class on pools of small maps and deploy each without retraining on larger, unseen layouts and agent populations. Across six layouts with 10,000 agents, our policies achieve throughput comparable to or higher than the strongest evaluated baselines. The proposed method has an average execution time below 10 ms per step, more than an order of magnitude faster than several established planners.

查看此版本解读 →查看此版本原文 ↗