Skip to content

论文arXiv 2611.1298311 天前

CAFE框架实现无指令场景驱动机器人自主任务规划与闭环执行 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:CAFE: Closed-Loop Autonomous Task Formulation and Execution without Instance-Specific Instructions(arXiv 2611.12983)

核心内容 ​

针对机器人缺乏具体指令时的自主性挑战,研究提出CAFE(闭环自主任务构建与执行)框架。该系统无需预定义动作序列,即可在清洁、安全及收纳等类别下自主决策。核心创新在于维护包含物体、功能区及执行状态的在线记忆,并通过规则引导的Trigger-LLM触发任务。在12个AI2-THOR场景及三种骨干网络测试中,Tri-LLM架构的GCR得分全面超越基线,证实了基于证据积累的任务生成有效性。

技术架构采用Draft、Complete、Audit三阶段LLM流水线,分别负责目标提案、细节解析与依赖检查,执行结果实时回写记忆以支持纠错与后续发现。消融实验验证了审计、反馈及触发机制的关键作用。LimoCobot真机测试展示了存储、清理及抓取失败后的恢复能力,表明共享场景状态能有效连接高层任务发现与底层物理执行,为具身智能在非结构化环境中的长程自主作业提供了可复现的算法范式。

关键看点 ​

  • 三LLM协同架构:集成Draft、Complete、Audit模块,分别承担目标生成、执行细节补全与依赖校验,替代传统硬编码规划器
  • 在线状态记忆:动态维护物体、功能区及执行状态图谱,作为Trigger-LLM触发依据,支持基于历史证据的任务自适应构建
  • 多维评测验证:在AI2-THOR 12场景中对比三种骨干网络,Tri-LLM在覆盖率、约束满足及执行成功率上均优于基线方法
  • 真机闭环验证:LimoCobot实测展示收纳、清理及抓取失败恢复全流程,验证共享状态机制对任务发现与物理执行的衔接能力

工程与产业观察 ​

CAFE证明了在无显式指令条件下,通过结构化记忆与多智能体LLM协作可实现鲁棒的场景驱动自主性。其“感知-记忆-规划-执行-修正”闭环设计,为解决具身智能长尾任务泛化难题提供了工程参考,尤其适用于家庭服务等非结构化环境的通用机器人部署。

英文摘要 (Abstract) ​

Scene-driven robot autonomy requires selecting tasks from observations as well as executing them. We study scene-driven task formulation and execution under predefined cleaning, safety, and storage/organization categories, without instance-specific goals or action sequences. CAFE maintains online memory of objects, functional zones, and execution states. A rule-guided Trigger-LLM admits formulation from accumulated evidence; Draft, Complete, and Audit propose goals, resolve execution details, and check dependencies. Outcomes update memory for correction and subsequent discovery. Evaluation in 12 AI2-THOR scenes compares three backbones and task-complexity tiers using coverage, constraint, and execution metrics. Tri-LLM records higher GCR scores than the evaluated baselines across three backbones and task-complexity tiers. Controls examine completion, auditing, zones, feedback, triggering, and prompt examples. LimoCobot traces illustrate storage, cleanup, and recovery from a failed grasp. The study examines how shared scene state connects task discovery to execution and outcome-driven revision.

作者:Mingjia Chen, Xuewen Zhao, Philip Dames, Qi Mao, Fei Xie, Jun Chen

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. arXiv 2611.12983:CAFE: Closed-Loop Autonomous Task Formulation and Execution without Instance-Specific Instructions ↗访问 2026-11-11