跳到正文
返回论文列表
cs.RO提交于 已译

EgoSwarm:从人类群体演示中学习去中心化机器人群体智能

EgoSwarm: Learning Decentralized Robot Swarm Intelligence from Human Swarm Demonstrations

Ziqiang Liu · Xitong Chen · Min Li · Yiqun Li · Zhuoping Yin

中文摘要

从个体具身智能迈向群体具身智能,是实现可扩展通用具身智能的关键一步。复杂任务常需多个智能体分工、协调交接并协同操作共享物体。现有机器人学习方法主要依赖个体演示,对智能体动作间依赖关系的监督有限,而大规模采集多机器人演示仍很困难。人类自我中心演示提供了可扩展的替代方案,但多参与者的分布式观测无法直接监督去中心化策略。EgoSwarm 是首个从人类群体演示中学习去中心化多机器人协作的机器人学习框架,通过无机器人多参与者自我中心系统采集超过 50 小时协作经验。引入 Swarm Interaction Packet (SIP) 与 Swarm Interaction Token (SIT),从分布式观测中恢复跨智能体交互关系,并将经验转化为与部署时信息约束一致的局部监督。进一步提出 SwarmVLA,将以局部观测、局部历史和有限邻居消息为条件的群体行为知识迁移到去中心化策略。SwarmVLA 在三类多机器人协作任务上平均成功率达 72.2%。项目页面:https://xitong-c.github.io/EgoSwarm_Page/。

关键要点

  1. 01问题:个体演示对智能体间动作依赖监督不足,多机器人演示难以规模化采集。
  2. 02方法:用无机器人多人自我中心系统采集 50+ 小时协作经验。
  3. 03方法:SIP 与 SIT 从分布式观测恢复跨智能体交互关系并转为局部监督。
  4. 04结果:SwarmVLA 在三类多机器人协作任务上平均成功率达 72.2%。
  5. 05局限:策略仅依赖局部观测、局部历史与有限邻居消息。

解读

尚无解读。

原始英文摘要

arXiv:2611.20100v1 Announce Type: new Abstract: Advancing from individual embodied intelligence to swarm embodied intelligence is a key step toward scalable, general-purpose embodied intelligence. Complex tasks often require multiple agents to divide responsibilities, coordinate handovers, and manipulate shared objects collaboratively. However, existing robot learning methods rely primarily on individual demonstrations, which provide limited supervision for modeling dependencies among agent actions, while collecting multi-robot demonstrations at scale remains difficult. Human egocentric demonstrations offer a scalable alternative, yet distributed observations from multiple participants cannot directly supervise decentralized policies. We introduce EgoSwarm, the first robot learning framework for learning decentralized multi-robot collaboration from human swarm demonstrations. EgoSwarm collects more than 50 hours of collaborative experience using a robot-free multi-person egocentric system. We introduce the Swarm Interaction Packet (SIP) and Swarm Interaction Token (SIT) to recover cross-agent interaction relations from distributed observations and transform this experience into local supervision consistent with deployment-time information constraints. We further propose SwarmVLA, which transfers knowledge of swarm behavior to decentralized policies conditioned on local observations, local history, and limited neighbor messages. SwarmVLA achieves a mean success rate of 72.2% across three categories of multi-robot collaborative tasks. Project page:https://xitong-c.github.io/EgoSwarm_Page/.

同方向论文 · cs.RO

查看全部 →

反向发现 · 这篇论文相关的工具与提示词

去交叉发现 →

关联工具 · 1

Google DeepMind76

Google DeepMind 发布 Gemini Robotics 2,为机器人带来全身智能控制

Google DeepMind 发布 Gemini Robotics 2,推出了三个模型:高级 VLA 模型可控制完整人形机器人从脚到指尖的运动,实现灵巧操作如系扣子;ER 2 推理模型可处理多步骤任务并协调多机器人协作;On-Device 2 模型可在几小时内适应全新机器人形态。

推荐理由:原文给出了三个模型的具体能力描述和可用入口,读者可以据此判断 Gemini Robotics 2 与前代及竞品的具体差异。